Safety index screening optimization method for gas turbine power plant risk assessment

Abstract

The invention discloses a safety index screening optimization method for gas turbine power plant risk assessment, and relates to the technical field of risk assessment, and the method comprises the steps: building an index architecture; obtaining a basic safety index; if the evaluation result of the current gas turbine power plant risk evaluation index exceeds an evaluation allowable error, re-screening the index; taking the safety index scheme of which the error value is smaller than the allowable evaluation error as a preliminary safety index scheme; obtaining an evaluation error fluctuation value; obtaining the index acquisition difficulty of the prepared safety index scheme; and obtaining an acquisition additional value, and taking the prepared safety index scheme with the minimum acquisition additional value as a target safety index scheme. The internal relation of the potential safety indexes is obtained through analysis, the potential safety indexes are preliminarily screened, the error value of the safety index scheme and gas turbine power plant risk assessment is calculated, and the assessment error fluctuation value is obtained, so that the assessment accuracy is ensured, and the operability of index collection is also ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and more particularly to a method for screening and optimizing safety indicators for hazard assessment of gas turbine power plants. Background Art

[0002] The operation of a power plant is a very complex systematic process, involving a variety of basic equipment. To ensure the stability of the operation process, it is necessary to assess the risks of the operation environment of each working link in the operation process of the power plant and eliminate various potential safety hazards in the bud.

[0003] During the assessment, the accuracy of the assessment has an important impact on the accuracy of the evaluation, and the assessment accuracy is mainly affected by the selection of indicators. However, there are many indicators involved in the assessment, and the relationships between the indicators are complex, making it difficult to accurately find better evaluation indicators. Summary of the Invention

[0004] To solve the above technical problems, a method for screening and optimizing safety indicators for hazard assessment of gas turbine power plants is provided, and the technical solution solves the problems raised in the above background art.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows: A method for screening and optimizing safety indicators for hazard assessment of gas turbine power plants, comprising: Based on the operation process of the gas turbine power plant, obtaining at least one potential safety indicator, and building an indicator framework based on the potential safety indicator; Based on the indicator framework, analyzing the internal relationships of the potential safety indicators, and preliminarily screening the potential safety indicators based on the internal relationships to obtain at least one basic safety indicator; Obtaining the allowable error of the hazard assessment of the gas turbine power plant. If the assessment result of the current hazard assessment indicator of the gas turbine power plant exceeds the allowable error of the assessment, then re-screen the indicators to obtain the weight setting of the basic safety indicators in the hazard assessment of the gas turbine power plant; Randomly combining at least one basic safety indicator to form at least one safety indicator scheme, calculating the error value between the safety indicator scheme and the hazard assessment of the gas turbine power plant, and taking the safety indicator scheme with an error value less than the allowable error of the assessment as the preliminary safety indicator scheme; Obtaining the fluctuation situation of the acquisition accuracy of the preliminary safety indicator scheme, and obtaining the evaluation error fluctuation value based on the fluctuation situation of the acquisition accuracy; Obtaining the index acquisition difficulty of the preliminary safety indicator scheme, where the index acquisition difficulty is composed of the acquisition time and the acquisition cost; Obtain the weights of the evaluation error fluctuation value and the difficulty of index collection. Based on the weights, accumulate the evaluation error fluctuation value and the difficulty of index collection to obtain the collection added value. Take the preliminary safety index plan with the smallest collection added value as the target safety index plan, and use the basic safety indicators in the target safety index plan to conduct a risk assessment of the gas turbine power plant.

[0006] Preferably, obtaining at least one potential safety index based on the operation process of the gas turbine power plant includes the following steps: During the gas transmission process, obtain the relevant equipment for gas transmission, and use at least one operating parameter of the relevant equipment for gas transmission as a potential safety index respectively; During the gas combustion process, obtain the relevant equipment for gas combustion control, and use at least one operating parameter of the relevant equipment for gas combustion control as a potential safety index respectively.

[0007] Preferably, building an index architecture based on the potential safety index includes the following steps: Obtain at least one historical event of abnormal operation of the gas turbine power plant, and count the abnormal amplitude of the potential safety index in the historical event; Calculate the variance of the ratio of the abnormal amplitudes of two potential safety indices as the correlation value. Take the potential safety indices as nodes, and connect line segments between the potential safety indices to form an index architecture. The length of the line segment between two potential safety indices is equal to the correlation value between the two potential safety indices.

[0008] Preferably, analyzing the internal relationship of the potential safety index based on the index architecture includes the following steps: Obtain the historical events where the potential safety index is abnormal and summarize them into a historical event set; Generate an i-index set of the potential safety index. The i-index set consists of the remaining i potential safety indices with the smallest distance from the potential safety index in the index architecture; Take the union of the historical event sets of the elements in the i-index set as the feature set. When the feature set contains the historical event set of the potential safety index corresponding to the i-index set, then take the i-index set as the preliminary i-index set; Take the preliminary i-index set with the smallest number of elements as the target i-index set, and take the elements in the target i-index set as the internal relationship of the potential safety index.

[0009] Preferably, the preliminary screening of the potential safety index based on the internal relationship includes the following steps: Summarize the potential safety indices using the target i-index set as the internal relationship to obtain an internal set, and pair the target i-index set with the internal set; Randomly form at least one combination of i-index sets using at least one target i-index set. When the union of the feature sets of the target i-index sets in the target i-index set combination contains all historical events, the i-index set combination is used as a preliminary i-index set combination; Take the union of the target i-index sets in the preliminary i-index set combination to obtain a first-level set, and take the union of the internal sets corresponding to the target i-index sets in the preliminary i-index set combination to obtain a second-level set; Take the difference set between the first-level set and the second-level set to obtain a third-level set, and select the potential safety indicators in the third-level set with the smallest number of elements as the basic safety indicators.

[0010] Preferably, the weight setting of the obtained basic safety indicators in the risk assessment of a gas turbine power plant includes the following steps: Obtain at least one basic safety indicator adopted in the current risk assessment of the gas turbine power plant as a characteristic basic safety indicator, and obtain the weight adopted by the characteristic basic safety indicator in the current risk assessment of the gas turbine power plant; Take the basic safety indicators other than the characteristic basic safety indicators as non-characteristic basic safety indicators; Take the mean value of the ratios of the abnormal amplitudes of the non-characteristic basic safety indicators and at least one characteristic basic safety indicator in historical events to obtain a first value; Take the mean value of the weights of at least one characteristic basic safety indicator to obtain a second value; Multiply the second value by the first value to obtain the weight of the non-characteristic basic safety indicator.

[0011] Preferably, the calculation of the error value between the safety indicator scheme and the risk assessment of the gas turbine power plant includes the following steps: Based on the occurrence results of historical events, obtain the losses caused by historical events as the risk values of historical events; Statistically obtain the historical values of the basic safety indicators in historical events, and use the basic safety indicators and the weight setting of the basic safety indicators in the safety indicator scheme to calculate the risk coefficient of historical events; Select the smallest risk value as the first target value, and select the smallest risk coefficient as the second target value; Take the ratio of the first target value to the second target value to obtain a regulation coefficient, and multiply the risk coefficient of historical events by the regulation coefficient to obtain the corrected risk coefficient of historical events; Take the absolute value of the difference between the corrected risk coefficient and the risk value of the same historical event, and obtain a local error. Accumulate at least one local error to obtain the error value between the safety indicator scheme and the risk assessment of the gas turbine power plant.

[0012] Preferably, obtaining the evaluation error fluctuation value based on the acquisition accuracy fluctuation condition includes the following steps: Obtain the minimum fluctuation value and the maximum fluctuation value of the acquisition accuracy fluctuation condition, and superimpose the historical values on the minimum fluctuation value and the maximum fluctuation value respectively to obtain the low point value and the high point value; Randomly use the low point value or the high point value to update and replace the historical value of the basic safety index. After each update and replacement, use the historical value of the basic safety index after the update and replacement to obtain a new error value between the safety index scheme and the risk assessment of the gas turbine power plant; Take the absolute value of the difference between the maximum value of the new error value and the original error value to obtain the first fluctuation difference, take the absolute value of the difference between the minimum value of the new error value and the original error value to obtain the second fluctuation difference, and take the maximum value of the first fluctuation difference and the second fluctuation difference as the evaluation error fluctuation value.

[0013] Preferably, obtaining the index acquisition difficulty of the preliminary safety index scheme includes the following steps: Obtain at least one acquisition step of the basic safety index in the preliminary safety index scheme, obtain the value obtained by multiplying the acquisition time and the acquisition cost of the acquisition step as the local acquisition value, and accumulate at least one local acquisition value to obtain the index acquisition difficulty of the preliminary safety index scheme.

[0014] Preferably, obtaining the weights of the evaluation error fluctuation value and the index acquisition difficulty includes the following steps: Use the analytic hierarchy process to obtain the weights of the evaluation error fluctuation value and the index acquisition difficulty.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: By analyzing the internal relationships of potential safety indicators, preliminarily screening potential safety indicators, calculating the error value between the safety indicator scheme and the risk assessment of the gas turbine power plant, and obtaining the evaluation error fluctuation value, it is possible to remove duplicates from the indicators by analyzing the internal relationships of the indicators, so that as few indicators as possible can be used for evaluation, thereby reducing the workload of the evaluation. At the same time, analyze the fluctuation of the acquisition accuracy during the acquisition of the indicators, so as to determine the impact of its acquisition accuracy fluctuation on the evaluation, and combine the acquisition difficulty. Thus, further screen out the scheme with the least comprehensive impact of the acquisition fluctuation and the acquisition difficulty on the evaluation, thereby ensuring both the accuracy of the evaluation and the operability of the indicator acquisition. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flow chart of the safety index screening and optimization method for gas turbine power plant risk assessment of the present invention; Figure 2 It is a schematic flow chart of obtaining at least one potential safety index based on the operation process of the gas turbine power plant of the present invention; Figure 3 Schematic flow chart of building an index framework based on potential safety indicators of the present invention; Figure 4 Schematic flow chart of analyzing the internal relationships of potential safety indicators based on the index framework of the present invention; Figure 5 Schematic flow chart of preliminarily screening potential safety indicators based on the internal relationships of the present invention; Figure 6 Schematic flow chart of obtaining the weight setting of basic safety indicators in the risk assessment of a gas turbine power plant according to the present invention; Figure 7 Schematic flow chart of calculating the error value between the safety index scheme and the risk assessment of a gas turbine power plant according to the present invention; Figure 8 Schematic flow chart of obtaining the evaluation error fluctuation value based on the fluctuation of the acquisition accuracy according to the present invention. Detailed implementation manners

[0017] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0018] Refer to Figure 1 As shown, a safety index screening and optimization method for the risk assessment of a gas turbine power plant includes: Based on the operation process of the gas turbine power plant, obtain at least one potential safety indicator, and build an index framework based on the potential safety indicator; Based on the index framework, analyze the internal relationships of the potential safety indicators, and based on the internal relationships, preliminarily screen the potential safety indicators to obtain at least one basic safety indicator; Obtain the evaluation allowable error of the gas turbine power plant risk assessment. If the evaluation result of the current gas turbine power plant risk assessment index exceeds the evaluation allowable error, re-screen the index to obtain the weight setting of the basic safety indicators in the gas turbine power plant risk assessment; Randomly combine at least one basic safety indicator to form at least one safety indicator scheme, calculate the error value between the safety indicator scheme and the gas turbine power plant risk assessment, and use the safety indicator scheme with an error value less than the evaluation allowable error as the preliminary safety indicator scheme; Obtain the acquisition accuracy fluctuation situation of the preliminary safety indicator scheme, and based on the acquisition accuracy fluctuation situation, obtain the evaluation error fluctuation value; Obtain the index acquisition difficulty of the preliminary safety indicator scheme, and the index acquisition difficulty is composed of the acquisition time and the acquisition cost; Obtain the weights of the evaluation error fluctuation value and the index acquisition difficulty. Based on the weights, accumulate the evaluation error fluctuation value and the index acquisition difficulty to obtain the acquisition added value. Take the preliminary safety index plan with the minimum acquisition added value as the target safety index plan, and use the basic safety indexes in the target safety index plan to conduct a risk assessment of the gas turbine power plant.

[0019] Accumulating the evaluation error fluctuation value and the index acquisition difficulty based on the weights means that the evaluation error fluctuation value and the index acquisition difficulty are multiplied by their corresponding weights respectively and then accumulated. In this solution, the screening of indexes is mainly considered from three aspects: the deduplication of indexes, the influence of the accuracy during index acquisition on the evaluation result, and the index acquisition difficulty. However, the focus is mainly on the first two aspects, and the acquisition difficulty is only considered incidentally. It is easy to know that there is a certain connection between indexes, so some indexes can be replaced by other indexes, that is, not all indexes need to participate in the evaluation to achieve a good evaluation effect. In addition, when collecting index data, different devices will be used for collection, and different devices will have fluctuations in different collection accuracies, which will lead to data fluctuations, and this part of the fluctuations will also cause fluctuations in the evaluation result. Therefore, in order to ensure that the selected indexes are better indexes, it is necessary to estimate the influence of the acquisition fluctuations on the evaluation result. Therefore, a series of steps will be set up later to achieve this goal.

[0020] Refer to Figure 2 As shown, based on the operation process of the gas turbine power plant, obtaining at least one potential safety index includes the following steps: During the gas transportation process, obtain the relevant equipment for gas transportation, and take at least one operating parameter of the relevant equipment for gas transportation as a potential safety index respectively; During the gas combustion process, obtain the relevant equipment for gas combustion control, and take at least one operating parameter of the relevant equipment for gas combustion control as a potential safety index respectively.

[0021] The gas turbine power plant mainly generates electricity through gas combustion. There are two parts with potential safety hazards. One is the gas transportation process, which may cause leakage and other situations. The other is the relevant indexes during the combustion process. For example, the indexes can be as follows: gas leakage value, gas pressure value, gas temperature value, gas pipeline corrosion coverage rate, gas leakage detection alarm response rate, gas pressure / temperature abnormal event times, gas pipeline corrosion detection coverage rate, emergency cut-off valve test qualification rate, combustion instability event times, flame detection system availability rate, burner maintenance timeliness rate, but it will vary according to the actual situation.

[0022] Refer to Figure 3As shown in the figure, based on potential safety indicators, the steps for building an indicator framework are as follows: Obtain at least one historical event in which the operation of the gas turbine power plant is abnormal. In the historical event, count the abnormal amplitude of the potential safety indicator; Calculate the variance of the ratio of the abnormal amplitudes of two potential safety indicators as the correlation value. Take the potential safety indicators as nodes and connect line segments between the potential safety indicators to form an indicator framework. The length of the line segment between two potential safety indicators is equal to the correlation value of the two potential safety indicators.

[0023] The construction of the indicator framework is mainly to visually display the relationship between potential safety indicators. The correlation value is an assessment of the degree of influence of potential safety indicators by historical events. Assuming that two potential safety indicators are strongly correlated, their degrees of influence by historical events are almost the same. Therefore, the ratio of the abnormal amplitudes of the two potential safety indicators in different historical events is almost the same. Therefore, the variance of the ratio is almost equal to 0. On the contrary, assuming that two potential safety indicators are not correlated, the variance of the ratio will be very large. Therefore, the correlation between two potential safety indicators can be evaluated according to the size of the correlation value. Since the length of the line segment between two potential safety indicators is set to be equal to the correlation value of the two potential safety indicators, the correlation between them can be evaluated through the distance between the two potential safety indicators in the indicator framework. The smaller the distance, the greater the correlation.

[0024] Refer to Figure 4 As shown in the figure, based on the indicator framework, the steps for analyzing the internal relationship of potential safety indicators are as follows: Obtain the historical events in which potential safety indicators are abnormal and summarize them into a historical event set; Generate an i-index set of potential safety indicators. The i-index set consists of the remaining i potential safety indicators with the smallest distance from the potential safety indicator in the indicator framework; Take the union of the historical event sets of the elements in the i-index set as the feature set. When the feature set contains the historical event set of the potential safety indicator corresponding to the i-index set, then take the i-index set as the preliminary i-index set; Take the preliminary i-index set with the smallest number of elements as the target i-index set, and take the elements in the target i-index set as the internal relationship of the potential safety indicator.

[0025] The elements in the target i-index set obtained by the above method can replace the potential safety indicator, because all the abnormalities that the potential safety indicator can reflect can be reflected by the elements in the target i-index set. Therefore, the elements in the target i-index set can be used to replace the potential safety indicator, thereby achieving the function of removing duplicates.

[0026] Refer to Figure 5As shown in the figure, based on the internal relationship, the preliminary screening of potential safety indicators includes the following steps: Summarize the potential safety indicators that use the target i indicator set as the internal relationship to obtain the internal set, and pair the target i indicator set with the internal set; Randomly form at least one i indicator set combination using at least one target i indicator set. When the union of the characteristic sets of the target i indicator sets in the target i indicator set combination contains all historical events, the i indicator set combination is used as the preliminary i indicator set combination; Take the union of the target i indicator sets in the preliminary i indicator set combination to obtain the first-level set, and take the union of the internal sets corresponding to the target i indicator sets in the preliminary i indicator set combination to obtain the second-level set; Take the difference set between the first-level set and the second-level set to obtain the third-level set, and select the potential safety indicators in the third-level set with the smallest number of elements as the basic safety indicators.

[0027] Due to the setting method of the preliminary i indicator set combination, the potential safety indicators in the third-level set formed by it can surely complete a comprehensive evaluation without abnormal omissions. However, in order to further reduce the workload, the number of indicators needs to be limited. Therefore, the potential safety indicators in the third-level set with the smallest number of elements are selected as the basic safety indicators.

[0028] Refer to Figure 6 As shown in the figure, the steps for obtaining the weight setting of the basic safety indicators in the risk assessment of a gas turbine power plant include the following: Obtain at least one basic safety indicator used in the current risk assessment of the gas turbine power plant as the characteristic basic safety indicator, and obtain the weight used for the characteristic basic safety indicator in the current risk assessment of the gas turbine power plant; Take the basic safety indicators other than the characteristic basic safety indicators as the non-characteristic basic safety indicators; Take the average value of the ratios of the abnormal amplitudes of the non-characteristic basic safety indicators and at least one characteristic basic safety indicator in historical events to obtain the first value; Take the average value of the weights of at least one characteristic basic safety indicator to obtain the second value; Multiply the second value by the first value to obtain the weight of the non-characteristic basic safety indicator.

[0029] Since the optimization is for indicators, an existing set of solutions must be adopted for the risk assessment of gas turbine power plants. In this set of solutions, weights are assigned to each indicator. However, these indicators do not fully cover the basic safety indicators. Therefore, as long as the weights of the included basic safety indicators are determined, the weights of the remaining basic safety indicators need to be further determined. Otherwise, the impact of the basic safety indicators on the assessment cannot be determined, and thus the screening of indicators cannot be carried out based on the results of the impact. The determination of the weights of the remaining basic safety indicators depends on the proportional relationship between the remaining basic safety indicators and the basic safety indicators with known weights.

[0030] Refer to Figure 7 As shown, calculating the error value between the safety indicator solution and the risk assessment of the gas turbine power plant includes the following steps: Based on the occurrence results of historical events, obtain the losses caused by historical events as the risk values of historical events; In historical events, count the historical values of basic safety indicators, and use the basic safety indicators and the weight settings of basic safety indicators in the safety indicator solution to calculate the risk coefficients of historical events; Select the smallest risk value as the first target value, and select the smallest risk coefficient as the second target value; Divide the first target value by the second target value to obtain a regulation coefficient, and multiply the risk coefficient of the historical event by the regulation coefficient to obtain the corrected risk coefficient of the historical event; Take the absolute value of the difference between the corrected risk coefficient and the risk value of the same historical event, and obtain the local error. Accumulate at least one local error to obtain the error value between the safety indicator solution and the risk assessment of the gas turbine power plant.

[0031] Under each set of safety indicator solutions, an assessment result of the risk of historical events will be generated. The actual results of historical events are known. Therefore, the assessment error can be determined by comparing the two.

[0032] Refer to Figure 8 As shown, obtaining the assessment error fluctuation value based on the fluctuation situation of the acquisition accuracy includes the following steps: Obtain the minimum fluctuation value and the maximum fluctuation value of the acquisition accuracy fluctuation situation, and add the historical values to the minimum fluctuation value and the maximum fluctuation value respectively to obtain the low point value and the high point value; Randomly use the low point value or the high point value to update and replace the historical values of the basic safety indicators. After each update and replacement, use the updated and replaced historical values of the basic safety indicators to obtain the new error value between the safety indicator solution and the risk assessment of the gas turbine power plant; Take the absolute value of the difference between the maximum value of the new error value and the original error value to obtain the first fluctuation difference. Take the absolute value of the difference between the minimum value of the new error value and the original error value to obtain the second fluctuation difference. Take the maximum value of the first fluctuation difference and the second fluctuation difference as the evaluation error fluctuation value.

[0033] Here, the minimum fluctuation value is negative and the maximum fluctuation value is positive. When each indicator is collected, its fluctuation may take the minimum fluctuation value or the maximum fluctuation value. Therefore, there are many possibilities for the fluctuations of all indicators. For each case, an error fluctuation analysis is carried out, that is, randomly use the low-point value or the high-point value to update and replace the historical values of the basic safety indicators. After each update and replacement, use the updated and replaced historical values of the basic safety indicators to obtain the new error value between the safety indicator scheme and the risk assessment of the gas turbine power plant. The method used here is the same as the steps for calculating the error value between the safety indicator scheme and the risk assessment of the gas turbine power plant, except that the historical values of the basic safety indicators are replaced with the low-point value or the high-point value before calculation.

[0034] The steps for obtaining the collection difficulty of the indicators of the preliminary safety indicator scheme include the following: Obtain at least one collection step of the basic safety indicators in the preliminary safety indicator scheme, obtain the product of the collection time and the collection cost of the collection step as the local collection value, and accumulate at least one local collection value to obtain the collection difficulty of the indicators of the preliminary safety indicator scheme.

[0035] The steps for obtaining the weights of the evaluation error fluctuation value and the collection difficulty of the indicators include the following: Use the analytic hierarchy process to obtain the weights of the evaluation error fluctuation value and the collection difficulty of the indicators.

[0036] Furthermore, this solution also proposes a storage medium, on which a computer-readable program is stored. When the computer-readable program is called, it executes the above-mentioned safety indicator screening and optimization method for gas turbine power plant risk assessment.

[0037] It can be understood that the storage medium can be a magnetic medium, such as a floppy disk, a hard disk, a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid-state drive (SSD).

[0038] In summary, the advantages of the present invention are as follows: By analyzing the internal relationships of potential safety indicators, preliminarily screening the potential safety indicators, calculating the error value between the safety indicator scheme and the risk assessment of the gas turbine power plant, and obtaining the evaluation error fluctuation value, it is possible to remove duplicates from the indicators through the analysis of the internal relationships of the indicators, so as to use as few indicators as possible for evaluation, thereby reducing the workload of evaluation. At the same time, analyze the fluctuation of the acquisition accuracy during the acquisition of indicators, so as to determine the impact of the acquisition accuracy fluctuation on the evaluation, and combine the acquisition difficulty. Thus, further screen out the scheme with the least comprehensive impact of the acquisition fluctuation and acquisition difficulty on the evaluation, thereby ensuring both the accuracy of the evaluation and the operability of the indicator acquisition.

[0039] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection required by the present invention is defined by the appended claims and their equivalents.

Claims

1. A safety index screening and optimization method for the risk assessment of a gas turbine power plant, characterized in that, Including: Based on the operation process of the gas turbine power plant, obtain at least one potential safety index, and build an index framework based on the potential safety index; Based on the index framework, analyze the internal relationship of the potential safety index, and based on the internal relationship, preliminarily screen the potential safety index to obtain at least one basic safety index; Obtain the evaluation allowable error of the gas turbine power plant risk assessment. If the evaluation result of the current gas turbine power plant risk assessment index exceeds the evaluation allowable error, re-screen the index to obtain the weight setting of the basic safety index in the gas turbine power plant risk assessment; Randomly combine at least one basic safety index to form at least one safety index scheme, calculate the error value between the safety index scheme and the gas turbine power plant risk assessment, and use the safety index scheme with the error value less than the evaluation allowable error as the preliminary safety index scheme; Obtain the acquisition accuracy fluctuation of the preliminary safety index scheme, and based on the acquisition accuracy fluctuation, obtain the evaluation error fluctuation value; Obtain the index acquisition difficulty of the preliminary safety index scheme, where the index acquisition difficulty is composed of acquisition time and acquisition cost; Obtain the weights of the evaluation error fluctuation value and the index acquisition difficulty, and based on the weights, accumulate the evaluation error fluctuation value and the index acquisition difficulty to obtain the acquisition added value. Use the preliminary safety index scheme with the smallest acquisition added value as the target safety index scheme, and use the basic safety index in the target safety index scheme to conduct the gas turbine power plant risk assessment.

2. The safety index screening and optimization method for gas turbine power plant risk assessment according to claim 1, characterized in that The step of obtaining at least one potential safety index based on the operation process of the gas turbine power plant includes the following steps: During the gas transmission process, obtain the relevant equipment for gas transmission, and use at least one operating parameter of the relevant equipment for gas transmission as the potential safety index respectively; During the gas combustion process, obtain the relevant equipment for gas combustion control, and use at least one operating parameter of the relevant equipment for gas combustion control as the potential safety index respectively.

3. The safety index screening and optimization method for the risk assessment of a gas turbine power plant according to claim 2, wherein The step of building an index framework based on the potential safety index includes the following steps: Obtain at least one historical event in which the operation of the gas turbine power plant is abnormal, and count the abnormal amplitude of the potential safety index in the historical event; Calculate the variance of the ratio of the abnormal amplitudes of two potential safety indexes as the correlation value. Use the potential safety index as a node, connect line segments between the potential safety indexes to form an index framework, and the line segment length between two potential safety indexes is equal to the correlation value of the two potential safety indexes.

4. The safety index screening and optimization method for the risk assessment of a gas turbine power plant according to claim 3, wherein The step of analyzing the internal relationship of the potential safety index based on the index framework includes the following steps: Obtain the historical events in which the potential safety index is abnormal and summarize them into a historical event set; Generate an i-index set of the potential safety index, where the i-index set is composed of the remaining i potential safety indexes with the smallest distance from the potential safety index in the index framework; Use the union of the historical event sets of the elements in the i-index set as the feature set. When the feature set contains the historical event set of the potential safety index corresponding to the i-index set, use the i-index set as the preliminary i-index set; Use the preliminary i-index set with the smallest number of elements as the target i-index set, and use the elements in the target i-index set as the internal relationship of the potential safety index.

5. The safety index screening and optimization method for the risk assessment of a gas turbine power plant according to claim 4, wherein The preliminary screening of potential safety indicators based on internal relationships includes the following steps: Summarize the potential safety indicators using the target i indicator set as the internal relationship to obtain the internal set, and pair the target i indicator set with the internal set; Randomly form at least one i indicator set combination using at least one target i indicator set. When the union of the characteristic sets of the target i indicator sets in the target i indicator set combination contains all historical events, then use the i indicator set combination as the preliminary i indicator set combination; Take the union of the target i indicator sets in the preliminary i indicator set combination to obtain the first-level set, and take the union of the internal sets corresponding to the target i indicator sets in the preliminary i indicator set combination to obtain the second-level set; Take the difference set between the first-level set and the second-level set to obtain the third-level set, and select the potential safety indicators in the third-level set with the smallest number of elements as the basic safety indicators.

6. The safety index screening and optimization method for gas turbine power plant risk assessment according to claim 5, wherein The weight setting of the basic safety indicators in the risk assessment of gas turbine power plants includes the following steps: Obtain at least one basic safety indicator used in the current risk assessment of gas turbine power plants as the characteristic basic safety indicator, and obtain the weight adopted by the characteristic basic safety indicator in the current risk assessment of gas turbine power plants; Take the basic safety indicators other than the characteristic basic safety indicators as non-characteristic basic safety indicators; Take the average of the ratios of the abnormal amplitudes of the non-characteristic basic safety indicators to at least one characteristic basic safety indicator in historical events to obtain the first value; Take the average of the weights of at least one characteristic basic safety indicator to obtain the second value; Multiply the second value by the first value to obtain the weight of the non-characteristic basic safety indicator.

7. The safety index screening and optimization method for the risk assessment of a gas turbine power plant according to claim 6, characterized in that The calculation of the error value between the safety indicator scheme and the risk assessment of gas turbine power plants includes the following steps: Based on the occurrence results of historical events, obtain the losses caused by historical events as the risk values of historical events; Statistical the historical values of the basic safety indicators in historical events, and use the basic safety indicators and the weight settings of the basic safety indicators in the safety indicator scheme to calculate the risk coefficients of historical events; Select the smallest risk value as the first target value, and select the smallest risk coefficient as the second target value; Take the ratio of the first target value to the second target value to obtain the regulation coefficient, and multiply the risk coefficient of the historical event by the regulation coefficient to obtain the corrected risk coefficient of the historical event; Take the absolute value of the difference between the corrected risk coefficient and the risk value of the same historical event, and obtain the local error. Accumulate at least one local error to obtain the error value between the safety indicator scheme and the risk assessment of gas turbine power plants.

8. The safety index screening and optimization method for the risk assessment of a gas turbine power plant according to claim 7, characterized in that The obtaining of the evaluation error fluctuation value based on the acquisition accuracy fluctuation situation includes the following steps: Obtain the minimum fluctuation value and the maximum fluctuation value of the acquisition accuracy fluctuation situation, and superimpose the historical values with the minimum fluctuation value and the maximum fluctuation value respectively to obtain the low point value and the high point value; Randomly use the low point value or the high point value to update and replace the historical values of the basic safety indicators. After each update and replacement, use the updated and replaced historical values of the basic safety indicators to obtain the new error value between the safety indicator scheme and the risk assessment of gas turbine power plants; Take the absolute value of the difference between the maximum value of the new error value and the original error value to obtain the first fluctuation difference. Take the absolute value of the difference between the minimum value of the new error value and the original error value to obtain the second fluctuation difference. Take the maximum value of the first fluctuation difference and the second fluctuation difference as the evaluation error fluctuation value.

9. The safety index screening and optimization method for the risk assessment of a gas turbine power plant according to claim 8, characterized in that The index acquisition difficulty of the described scheme for obtaining the preliminary safety index includes the following steps: Obtain at least one acquisition step of the basic safety index in the scheme for obtaining the preliminary safety index. Obtain the value obtained by multiplying the acquisition time and the acquisition cost of the acquisition step as the local acquisition value. Accumulate at least one local acquisition value to obtain the index acquisition difficulty of the scheme for obtaining the preliminary safety index.

10. The safety index screening and optimization method for the risk assessment of a gas turbine power plant according to claim 9, characterized in that The described steps for obtaining the weights of the evaluation error fluctuation value and the index acquisition difficulty include the following: Use the analytic hierarchy process to obtain the weights of the evaluation error fluctuation value and the index acquisition difficulty.

Images