A method for optimizing power equipment operation and maintenance based on intelligent algorithm

Abstract

The present invention discloses a method for optimizing the operation and maintenance of electric equipment based on an intelligent algorithm, which is applied to the technical field of operation and maintenance of electric equipment. The method comprises the following steps: obtaining the operation data and status data of the electric equipment at multiple time nodes to obtain a sample data set; training the equipment status prediction network and testing it based on the test set; setting the simulation time, simulating the electric equipment, and obtaining the simulated operation data of each time node during the simulated operation; inputting the simulated operation data into the equipment status prediction network after the test, and obtaining the simulated status data of each time node; calculating the operation evaluation index of each time node of the electric equipment based on the simulated operation data; and customizing the operation and maintenance plan of the equipment according to the operation evaluation index and the simulated status data of each time node. The present invention customizes and optimizes the operation and maintenance plan by simulating the electric equipment to predict the future operation status of the electric equipment, thereby improving the operation and maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment operation and maintenance, and more specifically to a power equipment operation and maintenance optimization method based on an intelligent algorithm. Background Art

[0002] Various power equipment in the power system is an important device for maintaining the normal operation of the power system. The operation and maintenance plan of the power equipment includes data monitoring, on-site inspection, functional testing, equipment inspection, equipment maintenance and fault repair at different frequencies. Only by ensuring the normal operation and maintenance of the power equipment can the normal operation of the power system be guaranteed. Therefore, the operation and maintenance of the power equipment is very important, and a suitable operation and maintenance plan can accurately reduce the probability of power equipment failure. However, the current operation and maintenance plans are mainly divided into two categories: one is to carry out daily inspections and maintenance according to a fixed maintenance plan, and to perform temporary repairs when a failure occurs. This plan has the problem of unreasonable arrangement. If the frequency is low, it will not have an effective maintenance effect, and if the frequency is high, it will waste a lot of costs; the other plan is to customize the operation and maintenance plan according to the historical data of the power equipment, improve the maintenance of high-risk equipment, and reduce the maintenance of low-risk equipment. Although this plan can save costs, the historical data cannot fully reflect the risk of the power equipment, nor can it provide a reference for the operation and maintenance of the power equipment in the future, and the assessment of the risk level requires a lot of calculations. Therefore, how to provide a power equipment operation and maintenance optimization method based on an intelligent algorithm is a problem that technicians in this field need to solve urgently. Summary of the invention

[0003] In view of this, the present invention provides a method for optimizing the operation and maintenance of power equipment based on an intelligent algorithm. The method predicts the status data of the power equipment in the future operation process by establishing an equipment status prediction network, obtains the operation data through simulation operation, and customizes the operation and maintenance plan of the power equipment within a certain period of time in the future based on the above data.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A method for optimizing power equipment operation and maintenance based on an intelligent algorithm comprises the following steps:

[0006] S1. Obtain the operation data and status data of retired power equipment at multiple time nodes to obtain a sample data set;

[0007] S2, dividing the sample data set into a training set and a test set, training the device state prediction network with the training set and testing it based on the test set;

[0008] S3. Setting the working state of the power equipment to include normal operation, repairable faults, and unrepairable faults, and setting the probability of change of the working state of the power equipment, simulation time, and time nodes;

[0009] S4. For each time node, all working states are assigned to the power equipment to obtain a working state collection;

[0010] S5. Repeat S4 for all time nodes to obtain a collection of simulation operation data for all time nodes;

[0011] S6. Determine the weights of different states at the same time node in the collection according to the probability of change of the working state of the power equipment, and calculate the simulation operation data at the same time node;

[0012] S7, inputting the simulated operation data into the tested equipment state prediction network to obtain the simulated state data of each time node;

[0013] S8. Calculate the operation evaluation index of each time node of the power equipment to be optimized based on the simulated operation data, and customize the operation and maintenance plan of the equipment according to the operation evaluation index of each time node and the simulated status data.

[0014] Optionally, the operating data in S1 includes operating time, number of failures, failure interval time, number of maintenance times, repair time and probability of change in working status, and the status data of the power equipment is the operating status of the power equipment, including normal operation, risk of failure, minor failure and serious failure.

[0015] Optionally, S2 is specifically:

[0016] S21, build a device status prediction network and initialize network parameters;

[0017] S22, using the running data as input samples and the state data at the same time node as the expected label;

[0018] S23, output the prediction result, compare the prediction result with the expected label, and calculate the error value;

[0019] S24, adjusting the sample weight and bias, and re-outputting the prediction result until the error value is less than the preset standard value;

[0020] S25. Test the device status prediction network using a test set.

[0021] Optionally, S6 is specifically:

[0022] For the same time node, the simulation operation data collection of the power equipment is a collection of simulation operation data of all simulation operation results of the power equipment in the time period from the previous time node to this time node. The weight of each simulation operation result is determined according to the probability of change of the working state of the power equipment. This weight is used as the weight of the simulation operation data corresponding to the simulation operation result. Weighted operation is performed on all simulation operation data, and the simulation operation data collection is converted into weighted simulation operation data.

[0023] Optionally, S5 is specifically:

[0024]

[0025] Where P is the operation evaluation index, P1 is the total operation time, P2 is the normal operation time, P3 is the single average normal operation time, P4 is the mean repair time for failures, P5 is the failure probability, and P 1E , P 2E , P 3E , P 4E , P 5E are the preset expected values ​​of P1, P2, P3, P4, and P5 respectively, and W1, W2, W3, W4, and W5 are the weight values.

[0026] Optionally, S6 specifically includes: determining an operation and maintenance plan based on the simulation status data of each time node, the operation and maintenance plan including data monitoring, on-site inspections, functional testing, equipment inspection, equipment maintenance and fault repair, and determining the execution frequency and intensity of the operation and maintenance plan within this time period based on the operation evaluation indicators of each time node.

[0027] It can be seen from the above technical solution that, compared with the prior art, the present invention provides a method for optimizing the operation and maintenance of power equipment based on an intelligent algorithm, which has the following beneficial effects: the present invention predicts the status data of power equipment in the future operation process by establishing an equipment status prediction network, obtains the operation data by means of simulated operation, and customizes the operation and maintenance plan of power equipment within a certain period of time in the future based on the above data, so that the operation and maintenance plan is more in line with the future maintenance needs of the power equipment, saves the operation and maintenance costs and avoids the waste of resources while ensuring the operation and maintenance effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] 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 embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0029] Figure 1This is a flow chart of the power equipment operation and maintenance optimization method of the present invention. DETAILED DESCRIPTION

[0030] 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.

[0031] The embodiment of the present invention discloses a method for optimizing power equipment operation and maintenance based on an intelligent algorithm, such as Figure 1 As shown, the following steps are included:

[0032] S1. Obtain the operation data and status data of retired power equipment at multiple time nodes to obtain a sample data set;

[0033] S2, dividing the sample data set into a training set and a test set, training the device state prediction network with the training set and testing it based on the test set;

[0034] S3. Setting the working state of the power equipment to include normal operation, repairable faults, and unrepairable faults, and setting the probability of change of the working state of the power equipment, simulation time, and time nodes;

[0035] S4. For each time node, all working states are assigned to the power equipment to obtain a working state collection;

[0036] S5. Repeat S4 for all time nodes to obtain a collection of simulation operation data for all time nodes;

[0037] S6. Determine the weights of different states at the same time node in the collection according to the probability of change of the working state of the power equipment, and calculate the simulation operation data at the same time node;

[0038] S7, inputting the simulated operation data into the tested equipment state prediction network to obtain the simulated state data of each time node;

[0039] S8. Calculate the operation evaluation index of each time node of the power equipment to be optimized based on the simulated operation data, and customize the operation and maintenance plan of the equipment according to the operation evaluation index of each time node and the simulated status data.

[0040] Furthermore, the operating data in S1 includes operating time, number of failures, failure interval time, number of maintenance times, repair time and probability of working status change, and the status data of the power equipment is the operating status of the power equipment, including normal operation, risk of failure, minor failure and serious failure.

[0041] In this embodiment, as a specific example, the state of the power equipment changes only once in the time period from the previous time node to this time node, and the length of the time node can be determined according to the average repair time of the power equipment. For example, the length of the time node is set to 2 days, and the length of the simulation time is set to 30 days;

[0042] The changing state of the power equipment between two time nodes is determined by the states of the two time nodes. The possible state changes of the power equipment include: continuous normal operation, conversion from normal operation to repairable fault, conversion from normal operation to unrepairable fault, continuous fault state, and conversion from repairable fault to normal operation. At the time node after a period of time, all possible states of the power equipment constitute the set of working states. During the simulation operation, the repair time of repairable faults adopts the average repair time of the power equipment.

[0043] Furthermore, S2 is specifically:

[0044] S21, build a device status prediction network and initialize network parameters;

[0045] S22, using the running data as input samples and the state data at the same time node as the expected label;

[0046] S23, output the prediction result, compare the prediction result with the expected label, and calculate the error value;

[0047] S24, adjusting the sample weight and bias, and re-outputting the prediction result until the error value is less than the preset standard value;

[0048] S25. Test the device status prediction network using a test set.

[0049] In this embodiment, as a specific example, the state prediction network adopts a BP neural network, and the error value is calculated as follows:

[0050]

[0051] In the formula, e is the error value, y is k is the prediction result of the kth input sample, y k * for y k The desired label.

[0052] Furthermore, S6 is specifically:

[0053] For the same time node, the simulation operation data collection of the power equipment is a collection of simulation operation data of all simulation operation results of the power equipment in the time period from the previous time node to this time node. The weight of each simulation operation result is determined according to the probability of change of the working state of the power equipment. This weight is used as the weight of the simulation operation data corresponding to the simulation operation result. Weighted operation is performed on all simulation operation data, and the simulation operation data collection is converted into weighted simulation operation data.

[0054] In this embodiment, based on the possible state changes of the power equipment, the duration of each state is different. As a specific example, if the state changes within a time period, the two states are located to last half of the time. For example, the possible state changes of an electric device are a set of continuous normal operation, conversion from normal operation to a repairable fault, and conversion from normal operation to an irreparable fault. The set of durations is 2 days of normal operation, 1 day of normal operation and 1 day of repairable fault, 1 day of normal operation and 1 day of irreparable fault. The weights of the three situations are determined by the hierarchical analysis method; for the next time node, the situation may be continuous normal operation, conversion from normal operation to a repairable fault, conversion from normal operation to an irreparable fault, continuous in a fault state, and conversion from a repairable fault to normal operation. The weights of the five situations are determined by the hierarchical analysis method; for the next time node, the situation may be continuous normal operation, conversion from normal operation to a repairable fault, conversion from normal operation to an irreparable fault, and continuous in a fault state. The weights of the four situations are determined by the hierarchical analysis method; all time nodes are calculated separately to obtain weighted simulated operation data.

[0055] Furthermore, S5 is specifically:

[0056]

[0057] Where P is the operation evaluation index, P1 is the total operation time, P2 is the normal operation time, P3 is the single average normal operation time, P4 is the mean repair time for failures, P5 is the failure probability, and P 1E , P 2E , P 3E , P 4E , P 5E are the preset expected values ​​of P1, P2, P3, P4, and P5 respectively, and W1, W2, W3, W4, and W5 are the weight values.

[0058] In this embodiment, W1, W2, W3, W4, and W5 are determined by the analytic hierarchy process.

[0059] Furthermore, S6 is specifically as follows: determining an operation and maintenance plan based on the simulation status data of each time node, the operation and maintenance plan including data monitoring, on-site inspection, functional testing, equipment inspection, equipment maintenance and fault repair, and determining the execution frequency and intensity of the operation and maintenance plan within this time period based on the operation evaluation indicators of each time node.

[0060] In this embodiment, as a specific example, the operation evaluation index is divided into 5 levels, and the 5 levels are divided into corresponding frequencies and intensities. Taking 2 days as a time node as an example, data monitoring is continuously monitored at all levels; the frequency of on-site inspections at levels 1-3 is once a day, and the frequency at levels 4-5 is twice a day; the frequency of functional testing and equipment inspections at levels 1-2 is once a day, the frequency at levels 3-4 is once every 2 days, and the frequency at level 5 is once every 4 days (two time nodes corresponding to the simulation state); the frequency of equipment maintenance at levels 1-2 is once every 2 days, and the frequency at levels 3-4 is once every 6 days (three time nodes corresponding to the simulation state). The frequency at level 5 is once every 10 days (5 time nodes corresponding to the simulation status), and combined with the equipment simulation status data of the corresponding time nodes, when the equipment simulation status data is normal operation, the operation and maintenance plan adopted is data monitoring and on-site inspection; when the equipment simulation status data is that there is a risk of failure, the operation and maintenance plan adopted is data monitoring, on-site inspection, and equipment maintenance; when the equipment simulation status data is that a minor failure occurs, the operation and maintenance plan adopted is data monitoring, on-site inspection, functional testing, equipment inspection, equipment maintenance and fault repair; when the equipment simulation status data is that a serious failure occurs, the operation and maintenance plan adopted is data monitoring and fault repair.

[0061] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0062] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing power equipment operation and maintenance based on intelligent algorithm, characterized in that: The following steps are involved: S1. Obtain the operation data and status data of retired power equipment at multiple time nodes to obtain a sample data set; S2, dividing the sample data set into a training set and a test set, training the device state prediction network with the training set and testing it based on the test set; S3. Setting the working state of the power equipment to include normal operation, repairable faults, and unrepairable faults, and setting the probability of change of the working state of the power equipment, simulation time, and time nodes; S4. For each time node, all working states are assigned to the power equipment to obtain a working state collection; S5. Repeat S4 for all time nodes to obtain a collection of simulation operation data for all time nodes; S6. Determine the weights of different states at the same time node in the collection according to the probability of change of the working state of the power equipment, and calculate the simulation operation data at the same time node; S7, inputting the simulated operation data into the tested equipment state prediction network to obtain the simulated state data of each time node; S8. Calculate the operation evaluation index of each time node of the power equipment to be optimized based on the simulated operation data, and customize the operation and maintenance plan of the equipment according to the operation evaluation index of each time node and the simulated state data; The operation data in S1 include operation time, number of failures, time between failures, number of repairs, repair time and probability of change in working status. The status data of power equipment are the operation status of power equipment, including normal operation, risk of failure, minor failure and major failure. S6 is specifically: For the same time node, the collection of simulated operation data of the power equipment is a collection of simulated operation data of all simulated operation results of the power equipment in the time period from the previous time node to this time node, and the weight of each simulated operation result is determined according to the probability of change of the working state of the power equipment, and this weight is used as the weight of the simulated operation data corresponding to the simulated operation result, and a weighted operation is performed on all simulated operation data, so as to convert the collection of simulated operation data into weighted simulated operation data; S8 is specifically: Where P is the operation evaluation index, P1 is the total operation time, P2 is the normal operation time, P3 is the single average normal operation time, P4 is the mean repair time for failures, P5 is the failure probability, and P 1E , P 2E , P 3E , P 4E , P 5E are the preset expected values ​​of P1, P2, P3, P4, and P5 respectively, and W1, W2, W3, W4, and W5 are the weight values.

2. According to claim 1, a method for optimizing power equipment operation and maintenance based on intelligent algorithm is characterized in that: S2 is specifically: S21, build a device status prediction network and initialize network parameters; S22, using the running data as the input sample and the state data at the same time node as the expected label; S23, output the prediction result, compare the prediction result with the expected label, and calculate the error value; S24, adjusting the sample weight and bias, and re-outputting the prediction result until the error value is less than the preset standard value; S25. Test the device status prediction network using a test set.

3. The method for optimizing power equipment operation and maintenance based on intelligent algorithm according to claim 1, characterized in that: S8 is specifically: Determine the operation and maintenance plan based on the simulation status data of each time node. The operation and maintenance plan includes data monitoring, on-site inspections, functional testing, equipment inspections, equipment maintenance and fault repairs. The execution frequency and intensity of the operation and maintenance plan within this time period are determined based on the operation evaluation indicators of each time node.

Images