A thermal power equipment status early warning evaluation method

Abstract

The invention discloses a thermal power equipment state early warning evaluation method, and relates to the technical field of thermal power equipment monitoring. The invention divides the thermal power equipment into a plurality of systems, and then selects the position of a collection device in each system and arranges it. When collecting data, effective data is selected according to the time resolution and collection frequency of the data, and whether the operation of each system is abnormal is analyzed. When at least two systems are abnormal, each faulty system is screened out according to the correlation between each data type between each system, and the fault level and emergency response plan of each faulty system are evaluated. At the same time, early warning prompts are given, so as to improve the accuracy and real-time performance of operation data collection, provide reliable data reference for subsequent early warning evaluation, confirm the faulty system according to the correlation between the systems, improve the accuracy of faulty equipment monitoring and early warning, and increase the effect of equipment early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power equipment monitoring, and in particular to a thermal power equipment state early warning evaluation method. Background Art

[0002] Thermal power equipment refers to various equipment used in thermal power plants to burn fossil fuels to generate electricity. Thermal power equipment operates under high temperature and high pressure conditions, which poses certain safety risks. Real-time monitoring of the operation of thermal power equipment and early warning when abnormalities are found in the equipment can avoid accidents such as explosions and fires, while reducing equipment downtime and improving power generation efficiency.

[0003] In the prior art, there is a method and system for intelligent operation of a thermal power unit, an electronic device and a storage medium disclosed in the application with publication number CN118094095A, which relates to the field of generator set control technology. The method includes: obtaining the unit operation data, preprocessing the unit operation data, and obtaining the preprocessed data; evaluating the unit operation status according to the preprocessed data to obtain the evaluation result; judging whether the abnormal value in the evaluation result meets the emergency alarm condition; if it meets, generating the corresponding emergency operation instruction and generating the alarm information according to the abnormal value; if it does not meet, generating the early warning information according to the abnormal value in the evaluation result, and sending it to the operator. The intelligent operation method of the thermal power unit provided by the present invention monitors and analyzes the unit operation data in real time by utilizing machine learning and pattern recognition technology, thereby realizing the accurate evaluation of the unit operation status and the ability to automatically generate emergency operation instructions and alarm information.

[0004] The above scheme has at least the following shortcomings: 1. The thermal power equipment contains many devices. When collecting the operating data of the equipment, the location of the collection equipment affects the accuracy of data collection. At the same time, the time resolution and collection frequency of the sensor when collecting data affect the real-time nature of the data. However, in the above scheme, when monitoring thermal power equipment, there is a lack of analysis of the layout location of the collection equipment and the time resolution and collection frequency of the data, which cannot improve the accuracy and real-time nature of the data and reduce the accuracy of subsequent equipment warnings.

[0005] 2. Thermal power equipment includes combustion system, steam-water system, electrical system and auxiliary system. Multiple devices in each system affect each other. When a problem occurs in a device in a system, it will have a negative impact on the operation of devices in other systems. Therefore, when the operating data of multiple devices are abnormal, it is impossible to directly determine whether all equipment failures or partial equipment failures affect other equipment. However, the above scheme lacks consideration of the mutual influence between equipment, cannot improve the accuracy of faulty equipment monitoring and early warning, and reduces the effectiveness of equipment early warning. Summary of the invention

[0006] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a thermal power equipment status early warning assessment method.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a thermal power equipment status early warning assessment method, comprising the following steps: S1, dividing the thermal power equipment into a combustion system, a steam-water system, an electrical system and an auxiliary system, and deploying a number of collection devices in each system to collect the operating data of each system.

[0008] S2. Analyze whether the operating status of each system is normal using the operating data of each system. When at least two systems have abnormal operating status, record the abnormal systems as marked systems, obtain the associated data and key data of each marked system, and screen out the faulty systems.

[0009] S3. Obtain the fault level of each faulty system and match the corresponding emergency response plan.

[0010] S4. Issue early warning for each faulty system and execute the corresponding emergency response plan for each faulty system.

[0011] The beneficial effects of the present invention are as follows: the present invention provides a thermal power equipment status early warning and evaluation method, the thermal power equipment is divided into multiple systems, and then the location of the acquisition equipment is selected and arranged in each system. When collecting data, effective data is selected according to the time resolution and acquisition frequency of the data, and then the operation of each system is analyzed to see if it is abnormal based on the effective operation data. When at least two system abnormalities occur, the faulty systems are screened out based on the correlation between the data types between the systems in the historical data, and the fault levels and emergency response plans of the faulty systems are evaluated. At the same time, early warning prompts are given to improve the accuracy and real-time performance of operation data collection, provide reliable data reference for subsequent early warning evaluation, confirm the faulty system based on the correlation between the systems, improve the accuracy of faulty equipment monitoring and early warning, and increase the effect of equipment early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0013] Figure 1 The present invention is a schematic flow chart of the steps for implementing the method. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] See also Figure 1 As shown, a method for early warning and evaluation of the status of thermal power equipment includes the following steps: S1, dividing the thermal power equipment into a combustion system, a steam-water system, an electrical system and an auxiliary system, and deploying several collection devices in each system to collect the operating data of each system.

[0016] It should be noted that the combustion system converts the chemical energy of the fuel into the thermal energy of steam, including equipment such as boilers, blowers, induced draft fans and air preheaters.

[0017] Steam-water system: converts the steam heat energy generated by the combustion system into mechanical energy and finally into electrical energy, including equipment such as steam turbines, condensers, deaerators and heaters.

[0018] Electrical system: Responsible for delivering the generated electrical energy to the grid, including equipment such as generators and main transformers.

[0019] Auxiliary systems include chemical water treatment system, cooling water system, dust removal system and desulfurization and denitrification system; Chemical water treatment system: used to treat raw water and prepare qualified boiler feed water, including activated carbon filters and water softeners and other equipment.

[0020] Cooling water system: provides cooling water for the turbine condenser to ensure the normal operation of the condenser, including cooling water pumps, cooling towers and other equipment.

[0021] Dust removal system: used to remove dust from flue gas and reduce air pollution, including equipment such as fans and dust collectors.

[0022] Desulfurization and denitrification system: used to remove sulfur dioxide and nitrogen oxides from flue gas to further reduce air pollution, including denitrification reactors and desulfurization pumps and other equipment.

[0023] In a specific embodiment, a number of collection devices are deployed in each system, and the specific process is as follows: the collection positions of each data type of operating data corresponding to each historical performance test in each system are obtained from the historical test data of the thermal power equipment, and the collection positions of each data type of operating data corresponding to each historical performance test in each system are clustered through position clustering technology to obtain the positions of each cluster center as the installation positions of each data type of operating data in each system.

[0024] From the historical inspection data of thermal power equipment, obtain the collection equipment model and effective collection rate of each data type corresponding to each historical performance inspection in each system, calculate the priority value of the collection equipment model of each data type, and select the collection equipment model corresponding to the maximum priority value as the collection equipment model of each data type.

[0025] Arrange the acquisition equipment models of each data type in each system at the corresponding installation positions to complete the layout of the acquisition equipment.

[0026] It should be noted that if the system is a combustion system, the data types include but are not limited to boiler furnace temperature, furnace pressure and air flow. The types of data collection equipment include but are not limited to temperature sensors, pressure sensors and air flow sensors.

[0027] In the above, the calculation process of the priority value of each data type collection device model is as follows: the number of collection times of each data type collection device model is counted, and divided by the total number of collection times of each data type collection device model, and then multiplied by the effective collection rate of each data type collection device model to obtain the priority value of each data type collection device model. The effective collection rate of each data type collection device model is obtained from the storage cloud of the device control center.

[0028] In another specific embodiment, the specific process of collecting the operating data of each system is as follows: S1-1, recording the collection devices of each data type as various types of collection devices, when various types of collection devices in each system are started, collecting data several times, and recording the time resolution and frequency of each data collection, recorded as and , where z represents the number of each system, a represents the number of each type of acquisition equipment, and b represents the number of each data acquisition. Both a and b are positive integers. z=c1, c2, c3, c4, where c1, c2, c3, c4 represent the combustion system, steam-water system, electrical system, and auxiliary system, respectively. Substitute them into the effective evaluation model expression: , output the data validity results of each system and each type of data collection equipment at each data collection , where , , They respectively represent the time resolution threshold, frequency threshold, and effective value threshold of type a acquisition device a in system z.

[0029] The time resolution and frequency of each valid data type in the historical records are obtained from the storage cloud of the device control center, and the minimum time resolution and frequency are selected as the time resolution threshold and the frequency threshold.

[0030] Time resolution describes the sampling density in time in the data, that is, the time interval between each data point or record. Higher time resolution means that the time interval between data points is small, which can more accurately capture the instantaneous changes of the device or system. Frequency is a specific manifestation of time resolution, which refers to the number of times or frequency of data collection per unit time. A high sampling frequency means a larger number of data points, which can provide more detailed and accurate time dynamic information. Time resolution and frequency directly affect the detail and real-time nature of the data, and are important factors that must be considered when selecting effective data.

[0031] It should be noted that the validity threshold is the critical value used to judge whether the data is valid, which is set by the management personnel of the thermal power equipment. For example, when the validity threshold is 2, The result is 1, and 1 is less than 2, which means that the data collected by the a-th type collection device in the z-th system at the b-th time is invalid.

[0032] S1-2. The validity result includes values ​​of 1 and 0. When the validity result is 1, it indicates that the data is valid. When the validity result is 0, it indicates that the data is invalid. All data with a validity result of 1 are selected as the valid data set, and the valid data in each system are combined as the operating data.

[0033] In the above, the number of valid data and the total number of data of each type of acquisition equipment in each system are counted, and the number of valid data is divided by the total number of data to obtain the efficiency of each type of acquisition equipment in each system. The models of each type of acquisition equipment are extracted to obtain the efficiency of each type of acquisition equipment model, and then uploaded to the storage cloud of the device control center for storage.

[0034] S2. Analyze whether the operating status of each system is normal using the operating data of each system. When at least two systems have abnormal operating status, record the abnormal systems as marked systems, obtain the associated data and key data of each marked system, and screen out the faulty systems.

[0035] In a specific embodiment, the operating status of the thermal power equipment is analyzed, and the specific process is as follows: the operating data safety interval of each system is obtained from the equipment control center, and the operating data of each system is compared with the operating data safety interval. If the operating data of at least one system is not within the operating data safety interval, it indicates that the operating status of the thermal power equipment is abnormal. If the operating data of all systems are within the operating data safety interval, it indicates that the operating status of the thermal power equipment is normal.

[0036] In another specific embodiment, the associated data and key data of each marking system are obtained, and the specific process is as follows: the system name corresponding to each single system failure and the abnormal data type set of other systems are obtained from the control center, thereby obtaining the abnormal data type set of other systems when each marking system fails, and the intersection of the abnormal data type sets of other systems when each marking system fails is selected as the key data type set of each marking system, based on the operating data of each system, the valid data set of each key data type of each marking system is extracted, and the valid data set of each key data type is averaged, and its value is taken as the value of each key data type, and then the value of each key data type in each marking system is used as the key data of each marking system.

[0037] The union of the abnormal data type sets of other systems at the time of each failure of each marking system is selected as the target data type set of each marking system. Based on the operating data of each system, the valid data set of each target data type of each marking system is extracted, and the mean is calculated and its value is taken as the value of each target data type. The data of each key data type is removed to obtain the value of each associated data class as the associated data of each marking system.

[0038] For example: when the steam-water system fails for the first time, the abnormal data types in the combustion system include steam temperature and feed water temperature; when the steam-water system fails for the second time, the abnormal data types in the combustion system include steam temperature and cooling water temperature, and the abnormal data type in the electrical system is generator speed. The intersection is the steam temperature in the combustion system, and the union is the steam temperature, feed water temperature, cooling water temperature of the combustion system and the generator speed of the electrical system. The key data of the steam-water system is the steam temperature in the combustion system. The associated data of the steam-water system is the union data minus the intersection data, including: the feed water temperature, cooling water temperature of the combustion system and the generator speed of the electrical system.

[0039] In a specific embodiment, the specific process of screening out the faulty systems is as follows: S2-1, classifying the marked systems into Class A systems, Class B systems and Class C systems.

[0040] In the above, the classification of each marking system is described, and the specific classification process is as follows: S2-11, the key data in each marking system are compared with each other. If there is at least one key data type in the key data corresponding to a marking system that is different from all the key data types in the key data corresponding to other marking systems, then the marking system is classified as a Class A system.

[0041] S2-12. Compare the key data and associated data in each marking system. If the key data types in the key data corresponding to a certain marking system are the same as the key data types in the key data corresponding to other marking systems, but at least one associated data type in the associated data is different from the associated data types in the associated data corresponding to other marking systems, then the marking system is classified as a Class B system; and the values ​​of the associated data types that are different from the associated data types in the associated data corresponding to other marking systems are recorded as feature data.

[0042] S2-13. If each key data type in the key data corresponding to a certain marking system is the same as each key data type in the key data corresponding to other marking systems, and each associated data type in the associated data is the same as each associated data type in the associated data corresponding to other marking systems, then the marking system is classified as a Class C system. In this way, each marking system is classified.

[0043] S2-2. Obtain the safety interval of each key data type in the key data corresponding to each Class A system from the equipment control center. If the values ​​of each key data type in the key data corresponding to a Class A system are all outside the corresponding safety interval, then the Class A system is a faulty system.

[0044] S2-3. Obtain the characteristic data of each Class B system, and obtain the safety interval of each key data type in the key data corresponding to each Class B system and the safety interval of each characteristic data type in the characteristic data from the equipment control center; if the values ​​of each key data type in the key data corresponding to a Class B system are all outside the corresponding safety interval, and there is at least one key data type in the key data that has a value outside the safety interval, then the Class B system is a faulty system.

[0045] S2-4. Obtain from the equipment control center the safety interval of each associated data type in the associated data corresponding to each Class C system and the safety interval of each key data type in the key data, and calculate the abnormal value of the associated data of each Class C system based on the associated data corresponding to each Class C system. The abnormal value includes values ​​of 1 and 0. When the abnormal value is 1, it indicates that the associated data is abnormal, and when the abnormal value is 0, it indicates that the associated data is normal. If the values ​​of each key data type in the key data corresponding to a Class C system are all outside the corresponding safety interval, and the abnormal value of the associated data is 1, it indicates that the Class C system is a faulty system.

[0046] In the above, the abnormal value of the associated data of each Class C system is calculated, and the specific process is as follows: the value of each associated data type in the associated data corresponding to each Class C system is compared with the corresponding safety interval to obtain the value difference of each associated data type, which is recorded as , r represents the number of each Class C system, f represents the number of each associated data type, and both r and f are positive integers.

[0047] Among them, if the value of a certain associated data type is smaller than the corresponding safety interval, the value difference of the associated data type is the lower limit of the safety interval minus the value of the associated data type; if the value of a certain associated data type is larger than the corresponding safety interval, the value difference of the associated data type is the value of the associated data type minus the upper limit of the safety interval.

[0048] Set the proportional coefficient of each associated data type in the associated data corresponding to each Class C system and record it as , substitute into the analytical formula , get the outlier value of the correlation data of the rth Class C system , where Indicates the preset safety value difference of the fth associated data type, represents the abnormal coefficient threshold of the associated data of the preset r-th Class C system, and g represents the number of associated data types.

[0049] In the above, the number of abnormalities of each associated data type in the associated data corresponding to each Class C system when each Class C system fails is counted from the equipment control center, and divided by the number of failures of each Class C system, to obtain the proportional coefficient of each associated data type in the associated data corresponding to each Class C system.

[0050] The safety numerical difference is the safety difference between the data and the safety interval, that is, within this difference range, the data is relatively safe. The safety numerical difference is set by the management personnel of the thermal power equipment based on experience. For example: assuming that the safety interval of the boiler furnace temperature in the combustion system is 900℃~1000℃, the safety numerical difference is 50, that is, it is relatively safe within 850℃~900℃ and 1000℃~1050℃.

[0051] The abnormal coefficient threshold represents the critical value for judging whether the associated data is abnormal. The calculation result is the abnormality coefficient of the correlation data of the rth Class C system. The abnormality coefficient of the correlation data corresponding to each failure of each Class C system is obtained from the equipment control center, and the minimum abnormality coefficient is selected as the abnormality coefficient threshold of each Class C system.

[0052] S2-5. Filter out each faulty system according to the analysis methods of S2-1 to S2-4.

[0053] S3. Obtain the fault level of each faulty system and match the corresponding emergency response plan.

[0054] In a specific embodiment, the specific process of obtaining the fault level of each fault system is as follows: comparing the value of each key data type in the key data corresponding to each fault system with the corresponding safety interval, obtaining the value difference of each key data type, recorded as , j represents the number of each fault system, k represents the number of each key data type, j and k are both positive integers; set the proportional coefficient of each key data type in the key data corresponding to each fault system, and record it as .

[0055] It should be noted that the process of obtaining the numerical difference of each key data type is the same as the process of obtaining the numerical difference of each associated data type in the corresponding associated data of each Class C system, and will not be repeated here.

[0056] The numerical differences of each key data type of the key data corresponding to each fault system at each fault are obtained from the equipment control center, and then divided by the average value of the numerical differences of each key data type of the key data corresponding to each fault, to obtain the numerical difference rate of each key data type of the key data corresponding to each fault system at each fault, and then the mean is calculated to obtain the average numerical difference rate of each key data type of the key data corresponding to each fault system, and the average numerical difference rate of each key data type of the key data corresponding to each fault system is divided by the sum of the average numerical difference rates of each key data type of the key data corresponding to each fault system to obtain the proportional coefficient of each key data type in the key data corresponding to each fault system.

[0057] Compare the values ​​of each associated data type in the associated data corresponding to each fault system with the corresponding safety interval to obtain the value difference of each associated data type in the associated data corresponding to each fault system, which is recorded as , set the proportional coefficient of each associated data type in the associated data corresponding to each fault system, and record it as .

[0058] It should be noted that the process of obtaining the numerical difference of each associated data type in the associated data corresponding to each fault system is the same as the process of obtaining the numerical difference of each associated data type in the associated data corresponding to each Class C system, which will not be repeated here. The process of setting the proportional coefficient of each associated data type in the associated data corresponding to each fault system is the same as the process of setting the proportional coefficient of each associated data type in the associated data corresponding to each Class C system, which will not be repeated here.

[0059] Set the key data ratio coefficient and the associated data ratio coefficient, respectively, as and , substitute it into the fault system analysis formula, calculate the fault coefficient of each fault system, compare the fault coefficient of each fault system with the fault coefficient interval corresponding to each fault level set in the equipment control system, and obtain the fault level of each fault system.

[0060] In the above, the average numerical difference rates of each key data type corresponding to each fault system are summed to obtain the average numerical difference rate of the key data, and according to the analysis process of the average numerical difference rate of the key data, the average numerical difference rate of the associated data is obtained, and the average numerical difference rate of the key data is divided by the sum of the average numerical difference rate of the key data and the average numerical difference rate of the associated data to obtain the key data proportional coefficient; the average numerical difference rate of the associated data is divided by the sum of the average numerical difference rate of the key data and the average numerical difference rate of the associated data to obtain the associated data proportional coefficient.

[0061] The fault system analysis formula is: , where represents the failure coefficient of the jth fault system, q represents the number of key data types, Indicates the security numeric difference of the preset k-th key data type.

[0062] S4. Issue early warning for each faulty system and execute the corresponding emergency response plan for each faulty system.

[0063] The equipment control center stores the emergency response plans corresponding to each fault system at each fault level, and extracts the emergency response plans of each fault system according to the fault level of each fault system.

[0064] For example: the fault levels of the combustion system include level one, level two and level three, among which level one > level two > level three. When the fault level of the combustion system is level three, the emergency response plan includes sending a maintenance reminder to the user end of the equipment management personnel; when the fault level of the combustion system is level two, the emergency response plan includes shutting down the equipment, and at the same time the alarm sounds an alarm and sending a level two fault warning prompt to the user end of the equipment management personnel; when the fault level of the combustion system is level one, the equipment shuts down, and at the same time the alarm sounds an alarm, and an evacuation reminder is sent to the user end of all personnel in the thermal power plant.

[0065] The embodiment of the present invention divides thermal power equipment into multiple systems, and then selects the location of the collection equipment in each system and arranges them. When collecting data, valid data is selected according to the time resolution and collection frequency of the data, and then the operation of each system is analyzed to see if it is abnormal based on the valid operation data. When at least two system abnormalities occur, the faulty systems are screened out based on the correlation between the data types between the systems in the historical data, and the fault levels and emergency response plans of the faulty systems are evaluated. At the same time, early warning prompts are given to improve the accuracy and real-time performance of operation data collection, provide reliable data reference for subsequent early warning evaluation, confirm the faulty system based on the correlation between the systems, improve the accuracy of faulty equipment monitoring and early warning, and increase the effect of equipment early warning.

[0066] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they shall all fall within the protection scope of the present invention.

Claims

1. A thermal power equipment status early warning assessment method, characterized in that: The steps include: S1. Divide the thermal power equipment into combustion system, steam-water system, electrical system and auxiliary system, and deploy several data collection devices in each system to collect the operation data of each system; S2. Analyze whether the operating status of each system is normal using the operating data of each system. If at least two systems have abnormal operating status, record the abnormal systems as marked systems, obtain the associated data and key data of each marked system, and screen out the faulty systems. S3. Obtain the fault level of each fault system and match the corresponding emergency response plan; S4. Issue early warning for each faulty system and execute the corresponding emergency response plan for each faulty system.

2. A thermal power equipment status early warning assessment method according to claim 1, characterized in that: The specific process of deploying several acquisition devices in each system is as follows: Obtain the collection locations of various data types of operating data corresponding to each historical performance test in each system from the historical test data of thermal power equipment, and use the location clustering technology to obtain the location of each cluster center as the installation location of each data type of operating data in each system; Obtain the data type collection equipment model and effective collection rate corresponding to each historical performance test in each system from the historical test data of thermal power equipment, calculate the priority value of the data type collection equipment model, and select the collection equipment model corresponding to the maximum priority value as the collection equipment model of each data type; Place the acquisition equipment models of each data type in each system at the corresponding installation positions to complete the layout of the acquisition equipment.

3. A thermal power equipment status early warning assessment method according to claim 2, characterized in that: The specific process of collecting the operating data of each system is as follows: S1-1. Record the data collection devices of each data type as various types of collection devices. When various types of collection devices in each system are started, collect data several times, and record the time resolution and frequency of each data collection, which are recorded as and , where z represents the number of each system, a represents the number of each type of acquisition equipment, and b represents the number of each data acquisition. Both a and b are positive integers. z=c1, c2, c3, c4, where c1, c2, c3, c4 represent the combustion system, steam-water system, electrical system, and auxiliary system, respectively. Substitute them into the effective evaluation model expression: , output the data validity results of each system and each type of data collection equipment at each data collection , where , , They represent the time resolution threshold, frequency threshold, and effective value threshold of the type a acquisition device a in the system z respectively; S1-2. The validity result includes values ​​of 1 and 0. When the validity result is 1, it indicates that the data is valid. When the validity result is 0, it indicates that the data is invalid. All data with a validity result of 1 are selected as the valid data set, and the valid data in each system are combined as the operating data.

4. A thermal power equipment status early warning assessment method according to claim 1, characterized in that: The specific process of analyzing the operating status of thermal power equipment is as follows: The operating data safety interval of each system is obtained from the equipment control center, and the operating data of each system is compared with the operating data safety interval. If the operating data of at least one system is not within the operating data safety interval, it indicates that the operating status of the thermal power equipment is abnormal. If the operating data of all systems are within the operating data safety interval, it indicates that the operating status of the thermal power equipment is normal.

5. A thermal power equipment status early warning assessment method according to claim 4, characterized in that: The specific process of obtaining the associated data and key data of each marking system is as follows: Obtain the system name corresponding to each single system failure and the abnormal data type set of other systems from the control center, thereby obtaining the abnormal data type set of other systems when each marking system fails, select the intersection of the abnormal data type sets of other systems when each marking system fails as the key data type set of each marking system, extract the valid data set of each key data type of each marking system based on the operating data of each system, and calculate the mean of the valid data set of each key data type, take its value as the value of each key data type, and then use the value of each key data type in each marking system as the key data of each marking system; The union of the abnormal data type sets of other systems at the time of each failure of each marking system is selected as the target data type set of each marking system. Based on the operating data of each system, the valid data set of each target data type of each marking system is extracted, and the mean is calculated and its value is taken as the value of each target data type. The data of each key data type is removed to obtain the value of each associated data class as the associated data of each marking system.

6. A thermal power equipment status early warning assessment method according to claim 5, characterized in that: The specific process of screening out the faulty systems is as follows: S2-1. Classify the marking systems into Class A systems, Class B systems and Class C systems; S2-2, obtaining the safety interval of each key data type in the key data corresponding to each Class A system from the equipment control center. If the values ​​of each key data type in the key data corresponding to a Class A system are all outside the corresponding safety interval, then the Class A system is a faulty system; S2-3, obtaining characteristic data of each Class B system, and obtaining from the device control center the safety interval of each key data type in the key data corresponding to each Class B system and the safety interval of each characteristic data type in the characteristic data; if the values ​​of each key data type in the key data corresponding to a Class B system are all outside the corresponding safety interval, and there is at least one key data type value in the key data that is outside the safety interval, then the Class B system is a faulty system; S2-4. Obtain the safety interval of each associated data type in the associated data corresponding to each Class C system and the safety interval of each key data type in the key data from the device control center, and calculate the abnormal value of the associated data of each Class C system according to the associated data corresponding to each Class C system. The abnormal value includes values ​​of 1 and 0. When the abnormal value is 1, it indicates that the associated data is abnormal, and when the abnormal value is 0, it indicates that the associated data is normal. If the values ​​of each key data type in the key data corresponding to a Class C system are all outside the corresponding safety interval, and the abnormal value of the associated data is 1, it indicates that the Class C system is a faulty system. S2-5. Filter out each faulty system according to the analysis methods of S2-1 to S2-4.

7. A thermal power equipment status early warning assessment method according to claim 6, characterized in that: The classification of each marking system is as follows: S2-11. Compare the key data in each marking system with each other. If at least one key data type in the key data corresponding to a marking system is different from all key data types in the key data corresponding to other marking systems, classify the marking system as a Class A system. S2-12. Compare the key data and associated data in each marking system. If each key data type in the key data corresponding to a marking system is the same as each key data type in the key data corresponding to other marking systems, and at least one associated data type in the associated data is different from each associated data type in the associated data corresponding to other marking systems, then classify the marking system as a Class B system. and recording the values ​​of each associated data type that is different from each associated data type in the associated data corresponding to each other marking system as feature data; S2-13. If each key data type in the key data corresponding to a certain marking system is the same as each key data type in the key data corresponding to other marking systems, and each associated data type in the associated data is the same as each associated data type in the associated data corresponding to other marking systems, then the marking system is classified as a Class C system. In this way, each marking system is classified.

8. A thermal power equipment status early warning assessment method according to claim 6, characterized in that: The specific process of calculating the abnormal value of the associated data of each Class C system is as follows: Compare the values ​​of each associated data type in the associated data corresponding to each Class C system with the corresponding safety interval to obtain the value difference of each associated data type, which is recorded as , r represents the number of each Class C system, f represents the number of each associated data type, and both r and f are positive integers; Set the proportional coefficient of each associated data type in the associated data corresponding to each Class C system, and record it as , substitute into the analytical formula , get the outlier value of the correlation data of the rth Class C system , where Indicates the preset safety value difference of the fth associated data type, represents the abnormal coefficient threshold of the associated data of the preset r-th Class C system, and g represents the number of associated data types.

9. A thermal power equipment status early warning assessment method according to claim 8, characterized in that: The specific process of obtaining the fault level of each fault system is as follows: Compare the values ​​of each key data type in the key data corresponding to each fault system with the corresponding safety interval to obtain the value difference of each key data type, which is recorded as , j represents the number of each fault system, k represents the number of each key data type, j and k are both positive integers; set the proportional coefficient of each key data type in the key data corresponding to each fault system, and record it as ; Compare the values ​​of each associated data type in the associated data corresponding to each fault system with the corresponding safety interval to obtain the value difference of each associated data type in the associated data corresponding to each fault system, which is recorded as , set the proportional coefficient of each associated data type in the associated data corresponding to each fault system, and record it as ; Set the key data ratio coefficient and the associated data ratio coefficient, respectively, as and , substitute it into the fault system analysis formula, calculate the fault coefficient of each fault system, compare the fault coefficient of each fault system with the fault coefficient interval corresponding to each fault level set in the equipment control system, and obtain the fault level of each fault system.

10. A thermal power equipment status early warning assessment method according to claim 9, characterized in that: The fault system analysis formula is: , where represents the failure coefficient of the jth fault system, q represents the number of key data types, Indicates the security numeric difference of the preset k-th key data type.

Images