阅读说明：本技术 一种火力发电机组实时健康度评价方法及系统 (Method and system for evaluating real-time health degree of thermal generator set ) 是由 苏烨 丁宁 蔡钧宇 孙坚栋 尹峰 张新胜 陈巍文 蒋薇 于 2021-07-28 设计创作，主要内容包括：本发明公开了一种火力发电机组实时健康度评价方法及系统。本发明采用的技术方案为：首先按照火力发电的工艺流程,将火电机组包含的全部设备分为机组级、系统级和设备级三个逻辑层级,然后确定与设备级各个设备健康度相关的状态变量及其基准值或基准值计算用的数学模型,并设置各级健康度分量的权重系数,随后自下而上、逐级地计算设备级、系统级的健康度；最后加权计算得到机组的实时健康度指数。由本发明得到的实时健康度指数是一个反映机组实时状态的综合性指标,通过该指标可以准确地把握机组的实时运行状态,提高火电机组的监盘效率、降低运行人员的工作强度。(The invention discloses a method and a system for evaluating the real-time health degree of a thermal generator set. The technical scheme adopted by the invention is as follows: firstly, dividing all equipment contained in a thermal power generating unit into three logic levels of a unit level, a system level and an equipment level according to a thermal power generation process flow, then determining a state variable related to the health degree of each equipment of the equipment level and a reference value thereof or a mathematical model for calculating the reference value, setting weight coefficients of health degree components of all levels, and then calculating the health degrees of the equipment level and the system level from bottom to top step by step; and finally, weighting and calculating to obtain the real-time health index of the unit. The real-time health index obtained by the invention is a comprehensive index reflecting the real-time state of the unit, and the real-time running state of the unit can be accurately grasped through the index, so that the monitoring efficiency of the thermal power unit is improved, and the working intensity of operators is reduced.)

1. A method for evaluating the real-time health degree of a thermal generator set is characterized by comprising the following steps:

step 1, dividing all equipment contained in one unit into three logic levels, namely a unit level, a system level and an equipment level according to a thermal power generation process flow;

step 2, determining all state variables related to the health degree of each device in the device level;

step 3, determining the upper limit H of the 1-level reference value of each state variable₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂Or a mathematical model for reference value calculation;

step 4, setting weight coefficients of the health degree components of the state variables in the total health degree of the equipment to which the state variables belong; determining a weight coefficient of the health degree component of each device in the total health degree of the system to which the health degree component belongs; determining the weight coefficient of the health degree component of each system in the total health degree of the unit;

and 5, calculating the health degree of each device at the device level: weighting and calculating four dimensions of deviation amount, change rate, overrun time and deviation amount integral of the state variable to obtain health degree component of the state variable; the health degree of each device is calculated by weighting the health degree components of the state variables contained in the device according to the weight coefficients;

and 6, calculating the health degree of the system level: the health degree of each system is calculated by weighting the health degree components of each device contained in the system according to a weight coefficient;

step 7, the health degree of the computer group level: and the health degree of the unit is calculated by weighting the health degree components of all the systems according to the weight coefficient.

2. The method according to claim 1, wherein in step 1, for some complex systems included in the plant, a subsystem level is added between a system level and an equipment level to make a logic relationship clearer.

3. The method according to claim 1, wherein in step 2, the state variables are selected to affect at least one of unit operation safety, operation cost, environmental protection and automatic control performance.

4. The method for evaluating the real-time health degree of the thermal generator set according to claim 1, wherein in the step 3, the reference values are divided into the following two types according to the characteristics of the state variables:

1) constant reference value

The reference value is a constant and cannot be changed along with the change of the operation condition, and is set according to a set threshold value or the specific operation requirement of the equipment;

2) variable reference value

Such reference values are variables that change with changes in operating conditions, and require a predetermined mathematical model for calculating the reference value and a dependent variable: firstly, determining one or more dependent variables related to the reference value through data correlation analysis, and then obtaining a mathematical model of the reference value changing along with the dependent variables by utilizing a modeling algorithm; when the health degree of a certain state variable is evaluated, the real-time measured value of the relevant dependent variable is input into the mathematical model, and the reference value of the state variable can be obtained through calculation.

5. The method for evaluating the real-time health degree of the thermal generator set according to claim 1, wherein in the step 4, the weight coefficients represent the importance degree of each health degree component in the total amount of the superior health degree to which the health degree component belongs, and the value is assigned according to the following three methods;

1) subjective empowerment method

Assigning a value to the weight coefficient of each health degree component;

2) objective weighting method for determining weight coefficient by entropy weight method

Determining objective weight coefficients of the components according to the variation degree of the health degree components;

3) combined weighting method combining subjective and objective functions

Firstly, the weight coefficient of each health degree component is determined by an entropy weight method, and then for some important state variables, equipment or systems, the corresponding weight coefficient is adjusted by a subjective weighting method.

6. The method for evaluating the real-time health degree of the thermal generator set according to claim 1, wherein the specific content in the step 5 is as follows:

1) calculating the health of all state variables contained in each device

Calculating normalized deviation value delta between each state variable and reference value_ij′_kIs provided with

In the formula, p_ijkValue of k-th state variable, H, contained in j-th device of i-th system_1ijkIs the upper limit of the level 1 reference value, L, of the state variable_1ijkIs the lower limit of the level 1 reference value, H, of the state variable_2ijkIs the upper limit of the 2-level reference value, L, of the state variable_2ijkA lower limit of a 2-level reference value of the state variable;

value of state variable p_ijkThe real-time measurement value is obtained from the DCS system of the unit, and is smoothed or taken as the average value of a plurality of continuous real-time measurement values in the latest period of time for eliminating noise;

for the constant type reference value, under different operation conditions, the value is a preset fixed value and does not need to be calculated; for the variable type reference value, the value of the variable type reference value changes along with the change of the operation condition, a dependent variable measurement value related to the reference value needs to be collected and input into a corresponding mathematical model, and the reference value is obtained through calculation;

calculating the normalized value v of the change rate of each state variable_i′_jkThe method comprises the following steps:

in the formula, v_ijkThe change rate of the kth state variable contained in the jth device of the ith system is as follows:

wherein T is the sampling period, V_ijkFor the upper limit of the change rate, nT represents the current sampling moment, and (n-1) T represents the last sampling moment;

calculating normalized value tau 'of each state variable overrun time'_ijkIs provided with

In the formula, τ_ijkOverrun time, T, of the kth state variable contained in the jth device of the ith system_ijkIs the upper limit of the overrun time;

fourthly, calculating normalized integral value chi 'of deviation between each state variable and reference value'_ijkIs provided with

In the formula, x_ijkIs the integral value, X, of the deviation between the kth state variable and the reference value contained in the jth equipment of the ith system_ijkIs the upper limit of the integral value;

weighting and calculating 4 dimensions of deviation, change rate, overrun time and deviation integral of the state variable to obtain the health degree of the state variable, i.e. the health degree

H_ijk＝(w_ijk1(1-δ′_ijk)+w_ijk2(1-δ′_ijk)(1-v′_ijk)+w_ijk3(1-τ′_ijk)+w_ijk4(1-χ′_ijk))×100 (6)

In the formula, w_ijk1、w_ijk2、w_ijk3And w_ijk4Weight coefficients which are respectively the deviation value, the change rate, the overrun time and the deviation value integral;

2) the health degree of each device at the device level is weighted and calculated by the health degree component of each state variable contained in the device level according to a weight coefficient, namely:

in the formula, H_ijHealth of the jth equipment of the ith system, H_ijkIs the health component of the k-th state variable, w, contained in the device_ijkWeight coefficient, N, for the kth state variable contained in the apparatus_ijkThe number of state variables contained for the device.

7. The method for evaluating the real-time health degree of the thermal generator set according to claim 1, wherein in the step 6, the health degree of each system at a system level is calculated according to the following formula:

in the formula, H_iIs the health of the ith system, H_ijIs the health component of the jth device included in the system, w_ijIs the weight coefficient of the j device, N_ijThe number of devices included in the system.

8. The method for evaluating the real-time health degree of the thermal generator set according to claim 7, wherein if the system is subordinate to a subsystem level, the health degree is calculated according to a formula (8), and the health degree components of the subsystems included in the system are weighted and calculated according to weight coefficients; if the system does not include a device or subsystem, the health degree is calculated as step 5, i.e. the health degree component of the state variable included in the system is weighted and calculated according to the weight coefficient.

9. The method for evaluating the real-time health degree of the thermal generator set according to claim 1, wherein in the step 7, a calculation formula of the health degree of the thermal generator set is as follows:

wherein H is the health of the unit, H_iIs the health component of the ith system, w_iAnd N is the weight coefficient of the ith system, and N is the number of the systems contained in the unit.

10. The utility model provides a thermal generator set real-time health degree evaluation system which characterized in that includes:

a classification unit: according to the technological process of thermal power generation, all equipment contained in one unit is divided into three logic levels, namely a unit level, a system level and an equipment level;

all state variable determination units: determining all state variables related to the health degree of each device at the device level;

a state variable reference value determination unit: determining the upper limit H of the reference value of level 1 of each state variable₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂Or a mathematical model for reference value calculation;

a weight coefficient unit: setting weight coefficients of the health degree components of the state variables in the total health degree of the equipment to which the state variables belong; determining a weight coefficient of the health degree component of each device in the total health degree of the system to which the health degree component belongs; determining the weight coefficient of the health degree component of each system in the total health degree of the unit;

an equipment-level health degree calculation unit: weighting and calculating four dimensions of deviation amount, change rate, overrun time and deviation amount integral of the state variable to obtain health degree component of the state variable; the health degree of each device is calculated by weighting the health degree components of the state variables contained in the device according to the weight coefficients;

a system-level health degree calculation unit: the health degree of each system is calculated by weighting the health degree components of each device contained in the system according to a weight coefficient;

unit level health degree computational element: and the health degree of the unit is calculated by weighting the health degree components of all the systems according to the weight coefficient.

Technical Field

The invention belongs to the technical field of power generation, and particularly relates to a method and a system for evaluating the real-time health degree of a thermal generator set.

Background

At present, thermal power generation (thermal power) is still the main electric energy production mode in China, and the installed capacity of thermal power generating units in China is 11.9 hundred million kilowatts and accounts for 59.2 percent of the total installed capacity by 2019 years according to the data display of the national statistical bureau; in 2019, when the thermal power generation amount is 51654.3 hundred million, the proportion of the thermal power generation amount accounts for 72.3 percent of the total power generation amount. With the gradual transformation of thermal power generation into an efficient, clean and environment-friendly power generation mode, the thermal power generation device still occupies an important position in the power industry of China in the long future.

The thermal power generating unit is a very large system, and comprises a plurality of auxiliary equipment such as a coal mill, an induced draft fan, a deaerator and the like besides three core main machines of a boiler, a steam turbine and a generator, and the process flow is complex, the production environment is severe, the working condition is complex and changeable, and a large amount of operators and maintainers are needed to ensure the long-term stable operation of the unit. The thermal power generating unit generally adopts a centralized control mode based on a Distributed Control System (DCS), each system and equipment are in an automatic control state, and operators monitor the operation condition of the thermal power generating unit through a human-computer interface, and execute operation according to parameter changes, external scheduling instructions and the like so as to enable the thermal power generating unit to operate safely, economically and environmentally. Because the thermal power generating unit equipment is large in scale and complex in production flow, operators need to continuously monitor mass operation parameters for 7 x 24 hours, and the working strength is extremely high. The traditional monitoring disc adopts a non-differential alarm-check mode, and operation of operators is prompted through alarm information. In consideration of safety, the upper and lower alarm limit values are set conservatively, alarm triggering is very frequent, effective alarm information is less, operating personnel needs to spend a great deal of energy to deal with various meaningless alarms, and monitoring efficiency is low. The existing monitoring mode only provides monitoring information about equipment operation parameters, and cannot provide comprehensive indexes reflecting the whole operation condition of a unit. In addition, although there are many factors affecting the operation of the thermal power generating unit, an operator can only monitor a few important pictures or operation parameters, so that local small degradation of unimportant systems and equipment cannot be found in time.

In recent years, it has become a consensus in the thermal power industry to construct an intelligent power plant, wherein an intelligent supervision board is an important application scenario of the intelligent power plant. The intelligent monitoring disk is used for deeply mining based on mass production data and high-quality operation experience, realizing deep detection and diagnosis of the unit state, providing efficient operation guidance for operators, and promoting high self-adaptation and optimization of an adjusting system.

The accurate evaluation of the real-time health degree of the thermal power generating unit is a basic condition for realizing intelligent monitoring, and the comprehensive index of the real-time health degree enables operators to master the comprehensive and accurate unit running state and provides a precondition for quickly and accurately executing various pre-control operations.

Disclosure of Invention

The invention aims to solve the technical problem of providing a method and a system for evaluating the real-time health degree of a thermal power generating unit, wherein the real-time health degree index obtained by calculation can enable operators to visually, comprehensively and accurately know and master the operating state of the thermal power generating unit without monitoring a large number of operating parameters.

Therefore, the invention adopts the following technical scheme: a method for evaluating the real-time health degree of a thermal generator set comprises the following steps:

step 1, dividing all equipment contained in one unit into three logic levels, namely a unit level, a system level and an equipment level according to a thermal power generation process flow;

step 2, determining all state variables related to the health degree of each device in the device level;

step 3, determining the upper limit H of the 1-level reference value of each state variable₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂Or a mathematical model for reference value calculation;

step 4, setting weight coefficients of the health degree components of the state variables in the total health degree of the equipment to which the state variables belong; determining a weight coefficient of the health degree component of each device in the total health degree of the system to which the health degree component belongs; determining the weight coefficient of the health degree component of each system in the total health degree of the unit;

and 5, calculating the health degree of each device at the device level: weighting and calculating four dimensions of deviation amount, change rate, overrun time and deviation amount integral of the state variable to obtain health degree component of the state variable; the health degree of each device is calculated by weighting the health degree components of the state variables contained in the device according to the weight coefficients;

and 6, calculating the health degree of the system level: the health degree of each system is calculated by weighting the health degree components of each device contained in the system according to a weight coefficient;

step 7, the health degree of the computer group level: and the health degree of the unit is calculated by weighting the health degree components of all the systems according to the weight coefficient.

Further, in step 1, for some complex systems included in the unit, a subsystem level is added between the system level and the equipment level, so that the logical relationship is clearer.

Further, in step 2, the state variable is selected to influence at least one of unit operation safety, operation cost, environmental protection and automatic control performance.

Further, in step 3, the reference values are classified into the following two types according to the characteristics of the state variables:

1) constant reference value

The reference value is a constant and cannot be changed along with the change of the operation condition, and is set according to a set threshold value or the specific operation requirement of the equipment;

2) variable reference value

Such reference values are variables that change with changes in operating conditions, and require a predetermined mathematical model for calculating the reference value and a dependent variable: firstly, determining one or more dependent variables related to the reference value through data correlation analysis, and then obtaining a mathematical model of the reference value changing along with the dependent variables by utilizing a modeling algorithm; when the health degree of a certain state variable is evaluated, the real-time measured value of the relevant dependent variable is input into the mathematical model, and the reference value of the state variable can be obtained through calculation.

Further, in step 4, the weight coefficients represent the importance degree of each health degree component in the total health degree of the superior health degree to which the weight coefficient belongs, and the weight coefficients are assigned according to the following three methods;

1) subjective empowerment method

Assigning a value to the weight coefficient of each health degree component;

2) objective weighting method for determining weight coefficient by entropy weight method

Determining objective weight coefficients of the components according to the variation degree of the health degree components;

3) combined weighting method combining subjective and objective functions

Firstly, the weight coefficient of each health degree component is determined by an entropy weight method, and then for some important state variables, equipment or systems, the corresponding weight coefficient is adjusted by a subjective weighting method.

Further, the specific content of step 5 is as follows:

1) calculating the health of all state variables contained in each device

Calculating deviation value normalized value delta 'between each state variable and reference value'_ijkIs provided with

In the formula, p_ijkValue of k-th state variable, H, contained in j-th device of i-th system_1ijkIs the upper limit of the level 1 reference value, L, of the state variable_1ijkIs the lower limit of the level 1 reference value, H, of the state variable_2ijkIs the upper limit of the 2-level reference value, L, of the state variable_2ijkA lower limit of a 2-level reference value of the state variable;

value of state variable p_ijkThe real-time measurement value is obtained from the DCS system of the unit, and is smoothed or taken as the average value of a plurality of continuous real-time measurement values in the latest period of time for eliminating noise;

for the constant type reference value, under different operation conditions, the value is a preset fixed value and does not need to be calculated; for the variable type reference value, the value of the variable type reference value changes along with the change of the operation condition, a dependent variable measurement value related to the reference value needs to be collected and input into a corresponding mathematical model, and the reference value is obtained through calculation;

calculating change rate normalized value v 'of each state variable'_ijkThe method comprises the following steps:

in the formula, v_ijkThe change rate of the kth state variable contained in the jth device of the ith system is as follows:

wherein T is the sampling period, V_ijkFor the upper limit of the change rate, nT represents the current sampling moment, and (n-1) T represents the last sampling moment;

calculating normalized value tau 'of each state variable overrun time'_ijkIs provided with

In the formula, τ_ijkOverrun time, T, of the kth state variable contained in the jth device of the ith system_ijkIs the upper limit of the overrun time;

fourthly, calculating normalized integral value chi 'of deviation between each state variable and reference value'_ijkIs provided with

In the formula, x_ijkIs the integral value, X, of the deviation between the kth state variable and the reference value contained in the jth equipment of the ith system_ijkIs the upper limit of the integral value;

weighting and calculating four dimensions of deviation, change rate, overrun time and deviation integral of the state variable to obtain the health degree of the state variable, i.e. the health degree

H_ijk＝(w_ijk1(1-δ′_ijk)+w_ijk2(1-δ′_ijk)(1-v′_ijk)+w_ijk3(1-τ′_ijk)+w_ijk4(1-χ′_ijk))×100 (6)

In the formula, w_ijk1、w_ijk2、w_ijk3And w_ijk4Weight coefficients which are respectively the deviation value, the change rate, the overrun time and the deviation value integral;

2) the health degree of each device at the device level is weighted and calculated by the health degree component of each state variable contained in the device level according to a weight coefficient, namely:

in the formula, H_ijHealth of the jth equipment of the ith system, H_ijkIs the health component of the k-th state variable, w, contained in the device_ijkWeight coefficient, N, for the kth state variable contained in the apparatus_ijkThe number of state variables contained for the device.

Further, in step 6, the health degree of each system at the system level is calculated as follows:

in the formula, H_iIs the health of the ith system, H_ijIs the health component of the jth device included in the system, w_ijIs the weight coefficient of the j device, N_ijThe number of devices included in the system.

Furthermore, if the system belongs to the subsystem level, the health degree calculation form is as shown in formula (8), and the health degree components of all the subsystems included in the system are weighted and calculated according to the weight coefficients; if the system does not include a device or subsystem, the health degree is calculated as step 5, i.e. the health degree component of the state variable included in the system is weighted and calculated according to the weight coefficient.

Further, in step 7, the calculation formula of the health degree of the unit is as follows:

wherein H is the health of the unit, H_iIs the health component of the ith system, w_iAnd N is the weight coefficient of the ith system, and N is the number of the systems contained in the unit.

The other technical scheme adopted by the invention is as follows: a method for evaluating the real-time health degree of a thermal generator set comprises the following steps:

a classification unit: according to the technological process of thermal power generation, all equipment contained in one unit is divided into three logic levels, namely a unit level, a system level and an equipment level;

all state variable determination units: determining all state variables related to the health degree of each device at the device level;

a state variable reference value determination unit: determining the upper limit H of the reference value of level 1 of each state variable₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂Or a mathematical model for reference value calculation;

a weight coefficient unit: setting weight coefficients of the health degree components of the state variables in the total health degree of the equipment to which the state variables belong; determining a weight coefficient of the health degree component of each device in the total health degree of the system to which the health degree component belongs; determining the weight coefficient of the health degree component of each system in the total health degree of the unit;

an equipment-level health degree calculation unit: weighting and calculating four dimensions of deviation amount, change rate, overrun time and deviation amount integral of the state variable to obtain health degree component of the state variable; the health degree of each device is calculated by weighting the health degree components of the state variables contained in the device according to the weight coefficients;

a system-level health degree calculation unit: the health degree of each system is calculated by weighting the health degree components of each device contained in the system according to a weight coefficient;

unit level health degree computational element: and the health degree of the unit is calculated by weighting the health degree components of all the systems according to the weight coefficient.

The real-time health degree index obtained by calculation is a comprehensive index reflecting the real-time running state of the unit, and operators can scientifically, comprehensively and accurately grasp the current running state of the unit by means of the index, so that the monitoring efficiency can be improved, and the working intensity can be reduced.

Drawings

Fig. 1 is a schematic flow chart of a method for evaluating the real-time health degree of a thermal power generating unit according to embodiment 1 of the present invention;

fig. 2 is a unit level schematic diagram of a thermal power generating unit real-time health degree evaluation algorithm according to embodiment 1 of the present invention;

fig. 3 is a graph showing changes in a real-time value of the outlet air temperature of the coal mill, an upper limit of a reference value of level 1, a lower limit of a reference value of level 1, an upper limit of a reference value of level 2, and a lower limit of a reference value of level 2 according to embodiment 1 of the present invention;

FIG. 4 is a graph showing the variation between the current of the coal mill and the input coal amount due to the dependent variable, and the upper limit of the reference value of level 1, the lower limit of the reference value of level 1, the upper limit of the reference value of level 2, the lower limit of the reference value of level 2 and the input coal amount calculated based on the unary linear regression model in example 1 of the present invention;

fig. 5 is a block diagram of a thermal power generating unit real-time health degree evaluation system according to embodiment 2 of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made with reference to the accompanying drawings. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.

Example 1

As shown in the attached figure 1, the method for evaluating the real-time health degree of the thermal generator set comprises the following steps:

step 1, as shown in fig. 2, according to the thermal power generation process flow, all the equipment included in one unit is divided into 3 logic levels, namely a unit level, a system level and an equipment level. For a certain 600MW supercritical thermal power generating unit, the system level comprises a water supply system, a heat recovery system, a condensed water system, a powder preparation system, a wind and smoke system, a steam turbine bypass system, a denitration system, a coal conveying system and the like; the lower part of the system level is an equipment level, for example, a water supply system comprises equipment such as an A, B side steam-driven water supply pump, an electric water supply pump and the like; a subsystem level can be set below the complex system, for example, the powder process system comprises 6 subsystems which are No. 1-6 powder process subsystems respectively, and each subsystem comprises a coal mill, a coal feeder, a cyclone separator and other equipment.

And 2, determining all state variables related to the health degree of each device at the device level. For the 600MW thermal power generating unit in the step 1, the state variables corresponding to the coal mill equipment comprise current, oil pressure, inlet-outlet differential pressure, outlet pressure, inlet air-powder temperature, outlet air-powder temperature, motor bearing temperature, coil temperature and the like.

Step 3, determining the upper limit H of the 1-level reference value of each state variable₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂Or a mathematical model for reference value calculation. The reference values are classified into the following two types according to the characteristics of the state variables:

(1) constant reference value

The reference value is a constant value and will not change with the change of the operation condition, and can be generally set according to the threshold value or the equipment operation requirement given by the manufacturer, the design institute, the electric department, and other units.

FIG. 3 shows the real-time value of the outlet air temperature of a coal mill and the upper limit H of the reference value of level 1₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂The reference value of the coal mill outlet air temperature is a constant.

(2) Variable reference value

Such reference values are variables that change with changes in operating conditions, requiring a mathematical model and dependent variables to be determined in advance to calculate the reference values. Firstly, one or more dependent variables related to the reference value are determined through data correlation analysis, and then a mathematical model of the change of the reference value along with the dependent variables is obtained by utilizing modeling algorithms such as linear regression and neural networks. When the health degree is evaluated, the real-time measured value of the relevant dependent variable is input into the mathematical model, and the reference value of the state variable can be obtained through calculation.

FIG. 4 shows the current and dependent variable input coal amount of a certain coal mill and the upper limit H of a 1-level reference value calculated based on a unary linear regression model₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂And the change curve of the input coal quantity.

Step 4, setting weight coefficients of the health degree components of the state variables in the total health degree of the equipment to which the state variables belong; determining a weight coefficient of the health degree component of each device in the total health degree of the system to which the health degree component belongs; and determining the weight coefficient of the health degree component of each system in the total health degree of the unit. The weighting coefficients of all levels can be assigned according to a subjective weighting method, an objective weighting method and a combined weighting method, and for a certain 600MW thermal power generating unit, some unimportant equipment can adopt the subjective weighting method, and important equipment and systems adopt the objective weighting method or the combined weighting method.

And 5, calculating the health degree of each device at the device level:

(1) calculating the health of all state variables contained in each device

Calculating deviation value normalized value delta 'between each state variable and reference value'_ijkIs provided with

In the formula, p_ijkValue of k-th state variable, H, contained in j-th device of i-th system_1ijkIs the upper limit of the level 1 reference value, L, of the state variable_1ijkIs a lower limit of a reference value of level 1, H_2ijkIs a level 2 reference value upper limit, L_2ijkIs a level 2 benchmark lower limit.

State variable p_ijkThe real-time measurement value of the state can be generally obtained from a set DCS system, and in order to eliminate noise, the real-time measurement value can be subjected to smoothing processing, or can be an average value of a plurality of continuous real-time measurement values in the latest period of time.

Calculating change rate normalized value v 'of each state variable'_ijkIs provided with

In the formula, v_ijkThe change rate of the kth state variable contained in the jth device of the ith system is

Where T is the sampling period, V_ijkFor the upper limit of the change rate, nT represents the current sampling moment, and (n-1) T represents the last sampling moment;

calculating normalized value tau 'of each state variable overrun time'_ijkIs provided with

In the formula, τ_ijkOverrun time, T, of the kth state variable contained in the jth device of the ith system_ijkThe upper limit of the overrun time.

Fourthly, calculating normalized integral value chi 'of deviation between each state variable and reference value'_ijkIs provided with

In the formula, x_ijkIs the integral value, X, of the deviation between the kth state variable and the reference value contained in the jth equipment of the ith system_ijkThe upper limit of the integration value.

Weighting and calculating 4 dimensions of deviation, change rate, overrun time and deviation integral of the state variable to obtain the health degree of the state variable, i.e. the health degree

H_ijk＝(w_ijk1(1-δ′_ijk)+w_ijk2(1-δ′_ijk)(1-v′_ijk)+w_ijk3(1-τ′_ijk)+w_ijk4(1-χ′_ijk))×100

(6)

In the formula, w_ijk1、w_ijk2、w_ijk3And w_ijk4Are the weight coefficients of the respective dimensions.

(2) The health degree of each equipment at the equipment level is calculated by weighting the health degree component of each state variable contained in the equipment according to a weight coefficient, i.e. the health degree of each equipment at the equipment level is calculated by weighting

In the formula, H_ijHealth of the jth equipment of the ith system, H_ijkIs the health component of the k-th state variable, w, contained in the device_ijkWeight coefficient for the kth state variable, N_ijkThe number of state variables contained for the device.

Step 6, the health degree of each system at the system level is weighted and calculated by the health degree component of each device contained in the system level according to the weight coefficient, namely

In the formula, H_iIs the health of the ith system, H_ijIs the health component of the jth device included in the system, w_ijIs the weight coefficient of the j device, N_ijThe number of devices included in the system.

If a certain system belongs to the subsystem level, the health degree calculation form is as shown in the formula (8), and the health degree components of all subsystems included in the system are weighted and calculated according to the weight coefficients. If the system does not include a device or subsystem, the health degree is calculated in step 5 by weighting the health degree components of the state variables included therein by weight coefficients.

And 7, weighting and calculating the health degree of the unit according to the health degree components of each system by weight coefficients, namely

Wherein H is the health of the unit, H_iIs the health component of the ith system, w_iAnd N is the weight coefficient of the ith system, and N is the number of the systems contained in the unit. The thermal power generating unit is a system which runs uninterruptedly, so that the real-time health index of the thermal power generating unit needs to be continuously evaluated, and after the calculation of the step is completed, the step 5 is returned to and repeatedly executed at certain intervals.

Example 2

The embodiment provides a real-time health degree evaluation system for a thermal generator set, which is composed of a grading unit, all state variable determination units, a state variable reference value determination unit, a weight coefficient unit, equipment health degree calculation units of equipment levels, a system-level health degree calculation unit and a unit-level health degree calculation unit, as shown in fig. 5.

A classification unit: according to the technological process of thermal power generation, all equipment contained in one unit is divided into three logic levels, namely a unit level, a system level and an equipment level;

all state variable determination units: determining all state variables related to the health degree of each device at the device level;

a state variable reference value determination unit: determining the upper limit H of the reference value of level 1 of each state variable₁Class 1 reference value lower limit L₁Class 2 reference value upper limit H₂Class 2 lower limit of reference value L₂Or a mathematical model for reference value calculation;

a weight coefficient unit: setting weight coefficients of the health degree components of the state variables in the total health degree of the equipment to which the state variables belong; determining a weight coefficient of the health degree component of each device in the total health degree of the system to which the health degree component belongs; determining the weight coefficient of the health degree component of each system in the total health degree of the unit;

an equipment-level health degree calculation unit: weighting and calculating four dimensions of deviation amount, change rate, overrun time and deviation amount integral of the state variable to obtain health degree component of the state variable; the health degree of each device is calculated by weighting the health degree components of the state variables contained in the device according to the weight coefficients;

a system-level health degree calculation unit: the health degree of each system is calculated by weighting the health degree components of each device contained in the system according to a weight coefficient;

unit level health degree computational element: and the health degree of the unit is calculated by weighting the health degree components of all the systems according to the weight coefficient.

Specifically, in the hierarchical unit, for some complex systems included in the unit, a subsystem level is added between a system level and an equipment level, so that the logical relationship is clearer.

Specifically, in all the state variable determination units, the state variable selects a state variable that affects at least one of unit operation safety, operation cost, environmental protection, and automatic control performance.

Specifically, in the state variable reference value determining unit, the reference values are classified into the following two types according to the characteristics of the state variable:

1) constant reference value

The reference value is a constant and cannot be changed along with the change of the operation condition, and is set according to a set threshold value or the specific operation requirement of the equipment;

2) variable reference value

Such reference values are variables that change with changes in operating conditions, and require a predetermined mathematical model for calculating the reference value and a dependent variable: firstly, determining one or more dependent variables related to the reference value through data correlation analysis, and then obtaining a mathematical model of the reference value changing along with the dependent variables by utilizing a modeling algorithm; when the health degree of a certain state variable is evaluated, the real-time measured value of the relevant dependent variable is input into the mathematical model, and the reference value of the state variable can be obtained through calculation.

Specifically, in the weight coefficient unit, the weight coefficient represents the importance degree of each health degree component in the total amount of the superior health degree to which the health degree component belongs, and is assigned according to the following three methods;

1) subjective empowerment method

Assigning a value to the weight coefficient of each health degree component;

2) objective weighting method for determining weight coefficient by entropy weight method

Determining objective weight coefficients of the components according to the variation degree of the health degree components;

3) combined weighting method combining subjective and objective functions

Firstly, the weight coefficient of each health degree component is determined by an entropy weight method, and then for some important state variables, equipment or systems, the corresponding weight coefficient is adjusted by a subjective weighting method.

Specifically, the specific contents of the device-level health degree calculation unit are as follows:

1) calculating the health of all state variables contained in each device

Calculating deviation value normalized value delta 'between each state variable and reference value'_ijkIs provided with

In the formula, p_ijkValue of k-th state variable, H, contained in j-th device of i-th system_1ijkIs the upper limit of the level 1 reference value, L, of the state variable_1ijkIs the lower limit of the level 1 reference value, H, of the state variable_2ijkIs the upper limit of the 2-level reference value, L, of the state variable_2ijkA lower limit of a 2-level reference value of the state variable;

value of state variable p_ijkThe real-time measurement value is obtained from the DCS system of the unit, and is smoothed or taken as the average value of a plurality of continuous real-time measurement values in the latest period of time for eliminating noise;

for the constant type reference value, under different operation conditions, the value is a preset fixed value and does not need to be calculated; for the variable type reference value, the value of the variable type reference value changes along with the change of the operation condition, a dependent variable measurement value related to the reference value needs to be collected and input into a corresponding mathematical model, and the reference value is obtained through calculation;

calculating change rate normalized value v 'of each state variable'_ijkThe method comprises the following steps:

in the formula, v_ijkThe change rate of the kth state variable contained in the jth device of the ith system is as follows:

wherein T is the sampling period, V_ijkFor the upper limit of the change rate, nT represents the current sampling moment, and (n-1) T represents the last sampling moment;

calculating normalized value tau 'of each state variable overrun time'_ijkIs provided with

In the formula, τ_ijkOverrun time, T, of the kth state variable contained in the jth device of the ith system_ijkIs the upper limit of the overrun time;

fourthly, calculating normalized integral value chi 'of deviation between each state variable and reference value'_ijkIs provided with

In the formula, x_ijkIs the integral value, X, of the deviation between the kth state variable and the reference value contained in the jth equipment of the ith system_ijkIs the upper limit of the integral value;

weighting and calculating four dimensions of deviation, change rate, overrun time and deviation integral of the state variable to obtain the health degree of the state variable, i.e. the health degree

H_ijk＝(w_ijk1(1-δ′_ijk)+w_ijk2(1-δ′_ijk)(1-v′_ijk)+w_ijk3(1-τ′_ijk)+w_ijk4(1-χ′_ijk))×100 (6)

In the formula, w_ijk1、w_ijk2、w_ijk3And w_ijk4Weight coefficients which are respectively the deviation value, the change rate, the overrun time and the deviation value integral;

2) the health degree of each device at the device level is weighted and calculated by the health degree component of each state variable contained in the device level according to a weight coefficient, namely:

in the formula, H_ijHealth of the jth equipment of the ith system, H_ijkIs the health component of the k-th state variable, w, contained in the device_ijkWeight coefficient, N, for the kth state variable contained in the apparatus_ijkThe number of state variables contained for the device.

Specifically, in the calculation unit of the system level health degree, a calculation formula of the health degree of each system of the system level is as follows:

in the formula, H_iIs the health of the ith system, H_ijIs the health component of the jth device included in the system, w_ijIs the weight coefficient of the j device, N_ijThe number of devices included in the system.

More specifically, if the system belongs to the subsystem level, the health degree calculation form is as shown in formula (8), and the health degree component of each subsystem included in the system is weighted and calculated according to the weight coefficient; if the system does not include a device or subsystem, the health degree is calculated as step 5, i.e. the health degree component of the state variable included in the system is weighted and calculated according to the weight coefficient.

Specifically, in the unit health calculation unit, a calculation formula of the unit health degree is as follows:

wherein H is the health of the unit, H_iIs the health component of the ith system, w_iAnd N is the weight coefficient of the ith system, and N is the number of the systems contained in the unit.

Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that various changes in the embodiments and/or modifications of the invention can be made, and equivalents and modifications of some features of the invention can be made without departing from the spirit and scope of the invention.