A withstand voltage test system and method for an insulated power cable

Abstract

The application discloses a withstand voltage test system and method for an insulated power cable, and particularly relates to the field of power cable insulation performance testing, which comprises obtaining object records, withstand voltage, holding time, discharge conditions and recovery intervals of a test cable, splitting the object records into node sequences according to the main body, terminal and joint, and generating an initial signature, performing first withstand voltage according to the withstand voltage and holding time, collecting the breakdown state, leakage current jump time and partial discharge starting time of each node, and writing the node number, collection time and collection result into a search node by using a Merkle-DAG node commitment pruning rule to generate a first commitment graph; the withstand voltage test system and method for the insulated power cable are characterized in that the first withstand voltage, discharge recovery and same-voltage re-withstand voltage are connected in series as a re-power transmission simulation test process, the recovery lag section is used to limit the re-voltage sampling boundary, and the first commitment graph is used to constrain the re-voltage abnormal early branch, so as to solve the problem that it is still difficult to identify the re-power transmission sensitive insulation weak point after single continuous withstand voltage.

Description

Technical Field

[0001] This invention relates to the field of power cable insulation performance testing technology, and more specifically, to a withstand voltage testing system and method for insulated power cables. Background Technology

[0002] In the acceptance and service testing of cross-linked polyethylene insulated power cables and cable accessories, the withstand voltage test is mainly used to confirm the dielectric withstand state of the line insulation under the specified test voltage. During the test, the voltage is generally applied to the specified withstand voltage value by the test power supply according to the voltage step-up procedure, and it is observed within the specified time whether breakdown, leakage current change or partial discharge abnormality occurs. When the test ends without any test stoppage, the conclusion of passing is formed. For newly laid lines, lines after maintenance, or lines after replacing intermediate joints, the test object has already completed terminal installation, joint fabrication, shield restoration, and sheath grounding. The on-site test time is limited by the power supply plan, and the test process cannot compensate for insufficient judgment by increasing the voltage, maintaining full voltage for a long time, or repeated impacts. At this time, the non-breakdown result obtained by a single continuous pressurization is difficult to reflect the insulation response when power is restored after a power outage. Micropores at the accessory interface, gaps in the insulation restoration layer, electric field concentration at the edge of the semiconductive layer peeling, or insufficient stress cone adhesion may temporarily show abnormal weakening during the first pressurization due to the accumulation of space charge and the redistribution of the local electric field. However, after the pressure is released, the residual charge is released, the interface electric field is re-established, and when the same withstand voltage is raised again, phenomena such as premature partial discharge initiation, premature leakage current transition, or tailing of absorption current after pressure release may occur. Therefore, a single continuous withstand voltage pass cannot fully demonstrate that the cross-linked polyethylene insulated power cable and cable accessories are still in a safe state under re-energization conditions. Therefore, the problem to be solved in this application is to identify the re-energized sensitive insulation weaknesses that were not exposed during the first continuous withstand voltage test, without increasing the specified withstand voltage or extending the full voltage holding time. Summary of the Invention

[0003] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a withstand voltage testing system and method for insulated power cables. By connecting the initial withstand voltage, discharge recovery, and re-withstand voltage at the same pressure in series as a re-energization simulation test process, and defining the re-voltage sampling boundary with the recovery lag segment and constraining the early branch of the re-voltage anomaly with the initial commitment diagram, the problems mentioned in the background art are solved.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for testing the withstand voltage of insulated power cables, comprising: S1. Obtain the object record, withstand voltage, holding time, discharge conditions and recovery interval of the cable under test, split the object record into node sequences according to the body, terminal and connector, and generate an initial signature; S2. Perform the first withstand voltage test according to the withstand voltage and holding time, collect the breakdown state, leakage current jump time and partial discharge start time of each node, and use the Merkle-DAG node commitment pruning rules to write the node number, collection time and collection results into the search node to generate the first commitment graph. S3. If the initial withstand voltage test fails to break down, perform discharge recovery according to the discharge conditions, collect the residual voltage value and the absorption current value, extract the recovery hysteresis segment according to the residual voltage drop and the absorption current drop, and write the initial signature, the initial commitment diagram and the recovery hysteresis segment into the status item to generate the recovery signature chain. S4. Based on the recovery signature chain, perform re-withstand voltage to bring the tested cable back to the same withstand voltage. Collect the re-withstand voltage leakage current jump time and the re-withstand voltage partial discharge start time. Generate a branch certificate by signing the node number, branch type, re-voltage lower limit time, re-withstand voltage abnormal time, first abnormal time and last status. Generate partial discharge early branch, leakage current early branch and absorption current tailing branch. S5. Perform branch pruning by recalculating branch certificates, comparing similar time points, and backtracking the initial commitment graph. Delete branches with broken signatures, out-of-bounds timing, and source mismatch, and output the remaining branches as re-energized sensitive insulation weakness records.

[0005] In a preferred embodiment, S1 includes: S1-1. Read the starting terminal position, ending terminal position, connector position and body length from the object record. Sort the starting terminal position, connector position and ending terminal position in ascending order of distance along the line. Bind the withstand voltage, holding time, discharge condition and recovery interval to the sorting result and output the position sequence. S1-2. Read the position sequence, write the starting and ending positions as terminal nodes, write the joint position as joint nodes, write the line segment between two adjacent positions as body nodes, and write the node number, node type, starting position, ending position and adjacent node number for each node, and output the node sequence. S1-3. Read the node sequence, withstand voltage, holding time, discharge condition and recovery interval. Generate the first node digest by hash digest operation using withstand voltage, holding time, discharge condition, recovery interval and the first node field. Then, starting from the second node, recursively generate the current node digest by hash digest operation using the previous node digest and the current node field. Output the last node digest as the initial signature.

[0006] In a preferred embodiment, S2 includes: S2-1. Read the node sequence, withstand voltage and holding time to perform the first withstand voltage test. Write the breakdown status, leakage current value and partial discharge pulse number according to the node number and the acquisition time to generate the first voltage sampling table. S2-2. Read the first voltage sampling table, calculate the difference between the current leakage current value and the previous leakage current value in ascending order of the sampling time of the same node, write the first sampling time when the difference is not zero as the leakage current jump time, write the first sampling time when the number of partial discharge pulses is not zero as the partial discharge start time, and generate the first abnormal entry time according to the earlier of the two times.

[0007] In a preferred embodiment, S2 further includes: S2-3. Read the first abnormal entry time, breakdown status and initial signature, write the node number, first abnormal entry time, leakage current jump time, partial discharge start time, previous acquisition time summary and adjacent node summary into the Merkle-DAG search node, and generate the node commitment summary through hash digest operation; S2-4. Read the Merkle-DAG search nodes. Group the search nodes with the same node number, breakdown state, leakage current jump time and partial discharge start time, and whose node commitment summary is derived from the summary of the same previous acquisition time and the summary of the same adjacent node into the same certificate node group. Retain the search node in the same certificate node group whose first abnormal entry time is earlier. Connect the retained search nodes according to the node sequence to generate the first commitment map as the benchmark for early re-withstand voltage determination.

[0008] In a preferred embodiment, S3 includes: S3-1. When the breakdown state in the initial commitment diagram is non-breakdown, read the discharge condition and recovery interval from the object record, break down the discharge condition into residual voltage limit and sampling period, and merge the node commitment summary of each search node in the initial commitment diagram level by level according to the node sequence to generate the initial commitment root signature and discharge task. S3-2. Perform discharge recovery according to the discharge task. Write the sampling time, residual voltage value and absorption current value according to the sampling period. Subtract the current residual voltage value from the previous residual voltage value to generate the residual voltage drop value. Subtract the current absorption current value from the previous absorption current value to generate the absorption current drop value. Output the recovery sampling table.

[0009] In a preferred embodiment, S3 further includes: S3-3. Read the recovery sampling table, write the first sampling time when the residual voltage value is not greater than the residual voltage limit as the discharge end point, calculate the difference between the residual voltage drop and the current absorption drop from the discharge end point according to the sampling time, write the sampling time when the difference changes from positive to negative as the recovery lag start point, write the sampling time when the difference changes from negative to positive as the recovery lag end point, and output the recovery lag segment. S3-4. Write the initial signature, the first commitment root signature, the recovery lag start point, the recovery lag end point, and the residual voltage drop and current drop within the recovery lag segment into the recovery status item according to the acquisition time. The current recovery status item is generated by the previous recovery signature and the current recovery status item. The recovery lag end point is output as the lower bound time of the re-withstand voltage advance, and the recovery signature chain is generated.

[0010] In a preferred embodiment, S4 includes: S4-1. Read the recovery lag segment, recovery interval, state item and last state signature from the recovery signature chain. Generate a recalculation signature by taking the item number, previous signature and recovery field in each state item in a fixed field order. When the recalculation signature is the same as the state signature in the same order in the recovery signature chain, write the end point of the recovery lag segment as the lower bound of the recalculation and generate a recalculation token. Otherwise, generate a recalculation blocking item. S4-2. After generating the re-voltage token, perform re-voltage withstand to bring the tested cable back to the same withstand voltage. Write the re-voltage leakage current value and the re-voltage partial discharge pulse number according to the acquisition time. Subtract the previous re-voltage leakage current value from the current re-voltage leakage current value to generate the re-voltage leakage current difference value, and output the re-voltage sampling table. S4-3. Read the complex voltage sampling table and the lower limit time of complex voltage. Use the lower limit window timing algorithm to delete complex voltage sampling items whose acquisition time is earlier than the lower limit time of complex voltage. Among the remaining complex voltage sampling items, take the first acquisition time when the complex voltage leakage current difference is not zero as the re-voltage withstand leakage current jump time. Take the first acquisition time when the re-voltage withstand partial discharge pulse number is not zero as the re-voltage withstand partial discharge start time. If no time is obtained, generate a complex voltage empty trigger item.

[0011] In a preferred embodiment, S4 further includes: S4-4. Read the re-voltage withstand leakage current jump time, the re-voltage withstand partial discharge start time, and the first commitment diagram. Use the DAG anchor point backtracking algorithm to search for the same node in the first commitment diagram by node number. Generate a recalculated node summary by combining the node number, the first leakage current jump time, the first partial discharge start time, the previous acquisition time summary, and the adjacent node summary in the same node search node. If the recalculated node summary is the same as the node commitment summary, generate the first voltage re-voltage time pair; otherwise, generate a backtracking mismatch item. S4-5. Read the initial voltage and re-voltage timing pair, the lower limit of the re-voltage, the recovery lag segment, and the last state signature. Use the dual-anchor certificate splitting algorithm to generate a branch certificate by combining the node number, branch type, lower limit of the re-voltage, re-voltage anomaly timing, first anomaly timing, and last state signature. When the re-voltage leakage current jump timing is later than the lower limit of the re-voltage but earlier than the first leakage current jump timing, write the branch certificate to the leakage current advance branch. When the re-voltage partial discharge start timing is later than the lower limit of the re-voltage but earlier than the first partial discharge start timing, write the branch certificate to the partial discharge advance branch. When the end of the recovery lag segment reaches the last moment of the recovery interval, write the branch certificate to the absorption current tailing branch. Write the remaining nodes to the non-advance branch.

[0012] In a preferred embodiment, S5 includes: S5-1: Read the partial discharge advance branch, leakage current advance branch, absorption current tailing branch and lower limit certificate. Generate a recalculation certificate by taking the node number, branch type, lower limit time of the re-voltage withstand voltage abnormality time, first similar time and last status signature in a fixed field order. If the recalculation certificate is the same as the lower limit certificate and the re-voltage withstand voltage abnormality time is after the lower limit time of the re-voltage withstand voltage, output the delimited branch. Otherwise, delete the branch and output the signature break item. S5-2. Read the bounded branches and adopt the same-time cross-pruning rule. Compare the start time of the re-withstand voltage partial discharge in the partial discharge advance branch with the start time of the first partial discharge. Compare the re-withstand voltage leakage current jump time in the leakage current advance branch with the first leakage current jump time. Compare the end time of the recovery lag segment in the absorption current tailing branch with the end time of the recovery interval. If the re-withstand voltage same-time is earlier than the first same-time or the end time of the recovery lag segment reaches the end time of the recovery interval, output the branch to be traced back. Otherwise, delete the branch and output the timing out-of-bounds item. S5-3. Read the branch to be traced back and the initial commitment graph. Using the commitment counterexample refeedback pruning rule, trace the node number of the branch to be traced back to the same node search node in the initial commitment graph. Then, recalculate the node commitment summary using the node number of the same node search node, the time of the first leakage current jump, the time of the first partial discharge start, the summary of the previous acquisition time, and the summary of adjacent nodes. If the recalculated node commitment summary is the same as the node commitment summary of the same node search node, output the re-energized sensitive insulation weakness record. Otherwise, delete the branch and write the node number, branch type, and deletion reason into the parent search node to form a counterexample summary.

[0013] In a preferred embodiment, a withstand voltage testing system for an insulated power cable includes: The node sequencing module is used to acquire the object record, withstand voltage, holding time, discharge conditions and recovery interval of the cable under test, split the object record into a node sequence according to the body, terminal and connector, and generate an initial signature; The first-voltage verification module is used to perform the first withstand voltage test according to the withstand voltage and holding time, collect the breakdown state, leakage current jump time and partial discharge start time of each node, and use the Merkle-DAG node commitment pruning rules to write the node number, acquisition time and acquisition results into the search node to generate the first commitment graph. The discharge verification module is used to perform discharge recovery according to the discharge conditions when the initial withstand voltage does not break down. It collects the residual voltage decay sequence and the absorption current fallback sequence, and uses the hash chain state signature rule to sequentially write the initial signature, the initial commitment graph digest and the discharge recovery data into the hash chain to generate the recovery signature chain. The re-voltage shunt module is used to perform re-voltage withstand based on the recovery signature chain, so that the tested cable is brought back to the same withstand voltage. It collects the re-voltage withstand leakage current jump time and the re-voltage withstand partial discharge start time, and generates a branch certificate by signing the node number, branch type, re-voltage lower limit time, re-voltage abnormal time, first abnormal time and last status signature. It also generates partial discharge early branch, leakage current early branch and absorption current tailing branch. The backtracking pruning module is used to backtrack and perform branch pruning based on branch certificate recalculation, comparison of similar time periods, and initial commitment graph. It deletes branches with broken signatures, out-of-bounds timing, and source mismatch, and outputs the remaining branches as re-energized sensitive insulation weakness records.

[0014] The technical effects and advantages of this invention are as follows: By connecting the initial withstand voltage, discharge recovery, and re-withstand voltage at the same voltage, the judgment of single non-breakdown is transformed into a comparison of the same time points of the initial and re-withstand voltages; without increasing the withstand voltage or extending the full voltage holding time, it is possible to identify re-energized sensitive insulation weaknesses that were not exposed during the initial withstand voltage test. By dividing the tested cable into a node sequence according to the body, terminal and connector, and binding the node field and withstand voltage parameter with the initial signature, the initial voltage sampling, discharge recovery and re-voltage judgment are all handled by the same node, reducing the confusion of abnormal attribution in long line and multiple accessory scenarios. The Merkle-DAG search node records the first abnormal entry time, leakage current jump time, and partial discharge start time, and generates the first commitment map, so that the first withstand voltage result becomes the benchmark for early judgment of re-withstand voltage, reducing the impact of occasional sampling changes on the judgment result; The recovery lag segment is extracted by the difference between the residual pressure reduction and the suction current reduction, and the recovery lag segment is written into the recovery signature chain, so that the discharge recovery process forms the lower limit of the pressure recovery time, reducing the data of the incomplete discharge stage from participating in the early branch judgment; By restoring the signature chain recalculation, the lower limit window of the voltage recovery time, and the generation of branch certificates, the partial discharge advance, leakage current advance, and absorption current tailing are all simultaneously constrained by the recovery boundary and the first voltage reference, thereby improving the correlation between the generation of voltage recovery branches and the risk of re-energization. By recalculating branch certificates, comparing similar time points, and backtracking the initial commitment graph, branches with broken signatures, out-of-bounds timing, and source mismatches are deleted, and only branches that can be backtracked to the initial commitment graph are retained, reducing the interference of occasional partial discharges and sampling jitter on vulnerability recording. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method steps of the present invention; Figure 2 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Refer to the instruction manual appendix Figure 1-2 The present invention provides a withstand voltage test method for insulated power cables, comprising: S1. Obtain the object record, withstand voltage, holding time, discharge conditions and recovery interval of the cable under test, split the object record into node sequences according to the body, terminal and connector, and generate an initial signature; In this embodiment, S1 is used to convert the line data and withstand voltage test parameters of the cable under test into a node sequence participating in the signature calculation, so that the subsequent initial withstand voltage, discharge recovery, and re-withstand voltage tests all use the same set of nodes as the objects for data acquisition, comparison, and tracing. The object record is generated from the test task sheet, as-built path data, and accessory installation records, and includes at least the cable under test number, starting and ending terminal positions, joint positions, body length, withstand voltage, holding time, discharge conditions, and recovery interval. Among them, the discharge conditions include residual voltage limits and sampling period, the withstand voltage and holding time are written by the test task, and the recovery interval is the duration between the discharge end and the re-withstand voltage start. This implementation process includes the following steps: S1-1. Read the starting terminal position, ending terminal position, connector position, body length, withstand voltage, holding time, discharge condition, and recovery interval from the object record. The starting terminal position is written as the zero point along the cable path, the ending terminal position is written as the body length, and the connector position is the distance from one or more connectors along the cable laying path to the starting terminal. If the ending terminal position is already written in the object record, compare the ending terminal position with the body length. If they are equal, use the ending terminal position; if they are not equal, use the body length as the ending terminal position and write the original ending terminal position into the object verification field. Then, arrange the starting terminal position, each connector position, and the ending terminal position in ascending order of distance along the cable path. If two connector positions have the same distance along the cable path, arrange them according to the connector number in the object record. After sorting, output the position sequence. Each item in the position sequence must include at least the position value and the position source type, where the position source type is either starting terminal, connector, or ending terminal. S1-2. Read the position sequence. Write the position items whose position source type is "starting terminal" or "ending terminal" as terminal nodes, and the position items whose position source type is "connector" as connector nodes. Write the line segment between two adjacent position items in the position sequence as body nodes. Write the starting and ending positions of the terminal nodes as terminal positions, the starting and ending positions of the connector nodes as connector positions, and the starting position of the body nodes as the position value of the previous position item and the ending position as the position value of the next position item. Write the node number, node type, starting position, ending position, and adjacent node number for the terminal nodes, connector nodes, and body nodes in the position sequence from the starting terminal to the ending terminal. Write the preceding node number of the first node in the sorted sequence as null, the following node number of the last node in the sorted sequence as null, and write the preceding and following node numbers for the intermediate nodes respectively. After writing all nodes, output the node sequence. The node sequence is used as the node index for subsequent withstand voltage sampling and summary backtracking. S1-3. Read the node sequence, withstand voltage, holding time, discharge condition, and recovery interval. Assemble the node field by combining the node number, node type, start position, end position, previous neighbor node number, and next neighbor node number in that order. Use null placeholders for null values ​​to ensure all nodes use the same field structure. For the first node in the node sequence, concatenate the withstand voltage, holding time, discharge condition, recovery interval, and the first node field in that order to form the first digest input string. Perform a hash digest operation on the first digest input string to obtain the first node digest. From the second node onwards, extract the digest of the previous node... The current node field is concatenated with the current node field in sequence to form the current digest input string. A hash digest operation is then performed on the current digest input string to obtain the current node digest. After recursively traversing the node sequence to the last node, the digest of the last node is output as the initial signature. It should be noted that the hash digest operation is implemented using a one-way hash function. The input fields are first converted into a field string in this order, and then a fixed-length digest value is generated by the one-way hash function. The first node digest is generated by the withstand voltage, holding time, discharge condition, recovery interval, and the first node field. The current node digest from the second node onwards is generated by the previous node digest and the current node field. Through the above processing, the line location, accessory location, and test parameters in the object record are converted into a node sequence with node number, location field, adjacency relationship, and recursive summary. The breakdown state, leakage current jump time, and partial discharge start time collected subsequently can be linked according to the node number, and the line structure and test parameters of the same cable under test can be traced through the initial signature. The initial signature is obtained by recursion from the node sequence and withstand voltage parameters. When the node location, accessory number, withstand voltage, holding time, discharge conditions, or recovery interval changes, the summary of the last node changes accordingly, so that the subsequent initial commitment diagram and recovery signature chain share the same data entry. In practical applications: If the starting and ending points of a cross-linked polyethylene insulated power cable are 0m, the ending point is 1200m, and the joint points are 420m and 810m, then the position sequence is arranged as 0m, 420m, 810m, and 1200m. 0m and 1200m are written as terminal nodes, 420m and 810m are written as joint nodes, and 0m to 420m, 420m to 810m, and 810m to 1200m are written as body nodes, then the node numbers and adjacent node numbers are written in the order along the line. Finally, the initial signature is generated by recursively combining the withstand voltage, holding time, discharge conditions, and recovery interval.

[0018] S2. Perform the first withstand voltage test according to the withstand voltage and holding time, collect the breakdown state, leakage current jump time and partial discharge start time of each node, and use the Merkle-DAG node commitment pruning rules to write the node number, collection time and collection results into the search node to generate the first commitment graph. In this embodiment, S2 is used to write the withstand voltage response of each node into a traceable commitment structure during the initial withstand voltage test, providing a first-voltage benchmark for subsequent re-withstand voltage assessment. The initial withstand voltage test takes the node sequence, withstand voltage, and holding time output by S1 as input, and connects the breakdown state, leakage current value, and partial discharge pulse number according to the node number. Then, the leakage current jump time, partial discharge start time, and first anomaly entry time are extracted from the sampled changes. Subsequently, a Merkle-DAG is constructed to search for nodes using the previous acquisition time summary and adjacent node summaries, so that the first voltage anomaly time of each node is simultaneously constrained by temporal precedence and spatial adjacency. This implementation process includes the following steps: S2-1. After reading the node sequence, withstand voltage, and holding time, the test power supply boosts the voltage according to the withstand voltage and enters the holding phase according to the holding time after reaching the withstand voltage. The acquisition device generates the acquisition time using the main clock of the test equipment and writes the breakdown status, leakage current value, and partial discharge pulse count according to the node number. The node number comes from the node sequence in S1. The breakdown status is "not broken down" or "broken down". When the test power supply protection trips, the test voltage cannot be held, or the test cable breakdown signal is triggered, it is written as "broken down". At other acquisition times, it is written as "not broken down". The leakage current value is output by the withstand voltage test power supply or the node current acquisition terminal and written as the quantized sampling code value. The partial discharge pulse count is counted by the partial discharge acquisition terminal in the same sampling period. When no partial discharge pulse is detected, it is written as zero. When detected, it is written as a positive integer. After completing the acquisition of the boosting and holding phases, the first voltage sampling table is generated according to the node number, acquisition time, breakdown status, leakage current value, and partial discharge pulse count. S2-2. After reading the initial voltage sampling table, first group by node number, then within the same node, arrange the leakage current value and partial discharge pulse count from earliest to latest according to the sampling time. For the leakage current value of the same node, subtract the previous leakage current value from the current leakage current value to obtain the leakage current difference, and write the sampling time when the first leakage current difference is not zero as the leakage current jump time. If the same node does not have a sampling time when the leakage current difference is not zero during the entire initial withstand voltage process, then the leakage current jump time is written as an empty value. For the partial discharge pulse count of the same node, write the sampling time when the first partial discharge pulse count is not zero as the partial discharge start time. If the partial discharge pulse count is always zero, then the partial discharge start time is written as an empty value. When neither the leakage current jump time nor the partial discharge start time is empty, take the earlier one and write it as the first abnormal entry time. When only one is not empty, take that one and write it as the first abnormal entry time. When both are empty, the first abnormal entry time is written as an empty value. S2-3. After reading the first abnormal entry time, breakdown status, and initial signature of each node, generate a Merkle-DAG search node for that node. Each Merkle-DAG search node includes a node number, acquisition time, breakdown status, first abnormal entry time, leakage current transition time, partial discharge start time, previous acquisition time summary, and adjacent node summary. Among them, the first acquisition time of the same node uses the aforementioned initial signature as the previous acquisition time summary, and the node commitment summary formed by the previous acquisition time is used as the previous acquisition time summary for subsequent acquisition times. The adjacent node summary is obtained by merging the previous neighbor node summary and the subsequent neighbor node summary in order and then performing a hash digest operation. If there is no previous neighbor node or subsequent neighbor node, write a null placeholder in the corresponding position. Then, concatenate the fields in the Merkle-DAG search node in a fixed order to form a node commitment input string, and perform a hash digest operation on the node commitment input string to generate a node commitment summary, so that the first voltage abnormal time of the node is simultaneously bound to the acquisition time predecessor and the line adjacency relationship. S2-4. After reading all Merkle-DAG search nodes, search nodes with the same node number, breakdown state, leakage current jump time, and partial discharge start time, and whose node commitment summary can be deduced from the summary of the same previous acquisition time and the summary of the same adjacent node, are grouped into a same-certification node group. Within the same-certification node group, the search node with the earliest first anomaly entry time is retained. If the first anomaly entry time is the same, the search node with the earliest acquisition time is retained. If the acquisition time is still the same, the search node with the earliest node commitment summary in lexicographical order is retained. After completing the same-certification node group processing, the retained search nodes are connected according to the node sequence, and the node commitment summary, the edge of the previous acquisition time summary, and the edge of the adjacent node summary are retained to generate the first commitment graph. The first commitment graph records the first leakage current jump time, the first partial discharge start time, and the first anomaly entry time of each node, which serves as the first voltage reference for judging the early anomaly during subsequent withstand voltage tests. Through the above processing, the initial withstand voltage process no longer generates only a set of sampled values, but instead forms an initial commitment graph with node number, acquisition time, initial voltage anomaly time, and summary commitment relationship; the leakage current jump time is obtained from the first change of the quantized sampled code value, the partial discharge start time is obtained from the first non-zero result of the partial discharge pulse count, and the first anomaly entry time is determined by the prior relationship between the two; the Merkle-DAG search node then binds this time with the summary of the previous acquisition time and the summary of adjacent nodes, so that subsequent withstand voltage results can only be traced back to the same node and the same initial voltage reference; In practical applications: During the initial withstand voltage holding period, at the 18th second, the leakage current value of a certain connector node changes relative to the 17th second. At the 23rd second, the number of partial discharge pulses changes from 0 to 2 for the first time. Therefore, the leakage current transition time of this connector node is written as 18s, the partial discharge start time is written as 23s, and the first abnormal entry time is written as 18s. Subsequently, the node number, 18s, 23s, breakdown state, previous acquisition time summary, and adjacent node summary of this node are written into the Merkle-DAG to search for nodes and generate a node commitment summary. Finally, it is written into the first commitment graph for the re-withstand voltage stage to determine whether the connector node has leakage current or premature partial discharge.

[0019] S3. If the initial withstand voltage test fails to break down, perform discharge recovery according to the discharge conditions, collect the residual voltage value and the absorption current value, extract the recovery hysteresis segment according to the residual voltage drop and the absorption current drop, and write the initial signature, the initial commitment diagram and the recovery hysteresis segment into the status item to generate the recovery signature chain. In this embodiment, S3 is used to convert the discharge recovery process into a recovery signature chain that can participate in re-voltage triggering and branch determination after the initial withstand voltage failure. The initial commitment graph is first used to confirm that the initial withstand voltage failure has not occurred, and the initial commitment root signature is generated from the node commitment digests of each search node. Then, discharge recovery is performed according to the discharge conditions, the residual voltage value and the absorbed current value are recorded, the recovery hysteresis segment is extracted by the difference between the residual voltage drop and the absorbed current drop, and the recovery hysteresis segment is written into the recovery status item so that the recovery hysteresis endpoint can be used as the time boundary of the re-voltage stage. This implementation process includes the following steps: S3-1. Read the breakdown status of each search node in the initial commitment diagram. If all search nodes are not broken, proceed to the discharge recovery process. If any search node is broken, do not generate a discharge task and write the breakdown node number into the withstand voltage test record. Read the discharge conditions and recovery interval from the object record. The discharge conditions are broken down into residual voltage limit and sampling period. The recovery interval is the time between the discharge end and the re-withstand voltage start. Then, read the node commitment summary of each search node in the initial commitment diagram and perform step-by-step merging according to the node sequence: two adjacent node commitment summaries are concatenated in the order of node number to generate the upper-level summary. If only one node commitment summary remains at the end of a certain level of merging, concatenate the node commitment summary with the null placeholder to generate the upper-level summary. When only one summary remains after step-by-step processing, write the summary as the initial commitment root signature. The discharge task consists of the tested cable number, residual voltage limit, sampling period, recovery interval, and initial commitment root signature. S3-2. Perform discharge recovery according to the discharge task. After the initial withstand voltage test, the discharge device connects the tested cable to the discharge circuit. The data acquisition device writes the sampling time, residual voltage value, and absorbed current value according to the sampling period. The residual voltage drop and absorbed current drop at the first sampling time are written as null values. From the second sampling time onwards, the residual voltage drop at the current sampling time is obtained by subtracting the current residual voltage value from the previous residual voltage value, and the absorbed current drop at the current sampling time is obtained by subtracting the current absorbed current value from the previous absorbed current value. The residual voltage drop and absorbed current drop retain their original signs: positive when the current value is lower than the previous value, zero when the current value is equal to the previous value, and negative when the current value is higher than the previous value. After the data acquisition is completed, a recovery sampling table is generated according to the sampling time, residual voltage value, absorbed current value, residual voltage drop, and absorbed current drop. S3-3. After reading the recovery sampling table, find the first sampling time in order of sampling time from earliest to latest where the residual voltage value is not greater than the residual voltage limit, and write this sampling time as the discharge end point; if there is no sampling time in the recovery sampling table where the residual voltage value is not greater than the residual voltage limit, write the last sampling time of the recovery interval as the discharge abnormal time, and do not generate a recovery lag segment; starting from the discharge end point, calculate the difference between the residual voltage drop and the current absorption drop item by item according to the sampling time; write the sampling time when the difference first turns from positive to negative as the recovery lag start point, and write the sampling time when the difference first turns from negative to positive after the recovery lag start point as the recovery lag end point; if there is no change from positive to negative, write both the recovery lag start point and the recovery lag end point as empty values; if the recovery lag start point has appeared but there is no change from negative to positive, write the last sampling time of the recovery interval as the recovery lag end point; the recovery lag segment consists of the recovery lag start point, the recovery lag end point, the sampling time within the segment, the residual voltage drop within the segment, the current absorption drop within the segment, and the difference within the segment; S3-4: Read the initial signature, the first commitment root signature, the recovery lag start point, the recovery lag end point, and the residual voltage drop and current absorption drop within the recovery lag segment, and generate recovery status items according to the acquisition time. Each recovery status item includes an item number, the previous signature, the acquisition time, the first commitment root signature, the recovery lag start point, the recovery lag end point, the current residual voltage drop, the current current absorption drop, and the current difference. The previous signature of the first recovery status item is taken from the initial signature, and the previous signature of subsequent recovery status items is taken from the recovery signature generated by the previous recovery status item. When generating the recovery signature, the previous signature and the fields in the current recovery status item are concatenated in order to form the recovery signature input string, and then a hash digest operation is performed on the recovery signature input string to obtain the current recovery signature. The current recovery status item does not contain the current recovery signature to avoid self-referencing. After recursively advancing to the last acquisition time of the recovery lag segment according to the acquisition time, all recovery status items and the last recovery signature are combined into a recovery signature chain. The recovery lag end point is written into the last recovery status item in the recovery signature chain for S4 to read as the re-voltage withstand advance lower bound time. Through the above processing, the discharge recovery process is no longer an auxiliary discharge step after the first withstand voltage test, but generates a recovery signature chain that can participate in the re-withstand voltage boundary determination; the residual voltage drop reflects the residual voltage decay, and the current absorption drop reflects the absorption current drop; the sign conversion of the difference between the two is used to truncate the recovery lag segment. After the recovery lag endpoint enters the recovery signature chain, the subsequent re-withstand voltage test will only read the abnormal trigger after this time boundary, thereby reducing the sampling disturbance of the incomplete discharge stage from entering the early branch; In practical applications: After the initial withstand voltage test fails to break down, the residual voltage limit in the discharge condition is 50V, and the sampling period is 1s; when the residual voltage value drops below 50V for the first time in the 12th second, the 12th second is written as the discharge end point; starting from the 12th second, the difference between the residual voltage drop and the current drawdown is calculated. If the difference changes from positive to negative in the 18th second and from negative to positive in the 31st second, then the 18th second is written as the recovery lag start point and the 31st second is written as the recovery lag end point. The residual voltage drop, current drawdown, and difference between the 18th and 31st seconds are written into the recovery status item, and the recovery signature chain is generated recursively.

[0020] S4. Based on the recovery signature chain, perform re-withstand voltage to bring the tested cable back to the same withstand voltage. Collect the re-withstand voltage leakage current jump time and the re-withstand voltage partial discharge start time. Generate a branch certificate by signing the node number, branch type, re-voltage lower limit time, re-withstand voltage abnormal time, first abnormal time and last status. Generate partial discharge early branch, leakage current early branch and absorption current tailing branch. In this embodiment, S4 is used to convert the recovery hysteresis segment in the recovery signature chain into the trigger boundary of the re-withstand voltage. After the tested cable is raised to the same withstand voltage again, the re-withstand voltage leakage current jump time and the re-withstand voltage partial discharge start time are extracted and compared with the first leakage current jump time and the first partial discharge start time in the initial commitment diagram. This process does not directly use full re-withstand voltage sampling. Instead, it first generates a re-voltage token by recalculating the recovery signature chain, then uses the end point of the recovery hysteresis segment as the lower limit time of the re-voltage to extract the re-voltage sampling item, and finally generates leakage current early branch, partial discharge early branch, absorption current tailing branch or non-early branch by DAG anchor point backtracking and dual anchor certificate shunting. This implementation process includes the following steps: S4-1. Read the recovery lag segment, recovery interval, status item, and last status signature from the recovery signature chain. Recalculate the signature of each status item in sequence according to the item number. Each status item includes an item number, the previous signature, and a recovery field. The recovery field includes the acquisition time, the first commitment root signature, the recovery lag start point, the recovery lag end point, the residual voltage reduction, the current absorption reduction, and the difference. During recalculation, concatenate the item number, the previous signature, and the recovery field in the field order when generating the recovery signature chain to obtain the recalculation input string. Then, perform a hash digest operation on the recalculation input string to generate the recalculation signature. Compare the recalculation signature with the status signatures of the same item number in the recovery signature chain item by item. When the recalculation signatures corresponding to all item numbers are the same as the status signatures in the same order in the recovery signature chain, read the end point of the recovery lag segment, write the end point of the recovery lag segment as the lower bound time of the recalculation, and generate a recalculation token. If the recalculation signature corresponding to any item number is different, do not perform re-voltage withstand, and write the corresponding item number, the previous signature, and the recalculation signature into the recalculation blocking item. S4-2. After generating the re-voltage token, the test power supply performs a re-voltage withstand test according to the withstand voltage in S1, so that the tested cable is brought back to the same withstand voltage. The acquisition device writes the node number, re-voltage leakage current value, and re-voltage partial discharge pulse count according to the acquisition time. The re-voltage leakage current value is written using the same quantization sampling method as the first withstand voltage test, and the re-voltage partial discharge pulse count is written using the same partial discharge counting method as the first withstand voltage test. For the same node, the current re-voltage leakage current value is subtracted from the previous re-voltage leakage current value to generate the re-voltage leakage current difference. If there is no previous re-voltage leakage current value at the first acquisition time, the re-voltage leakage current difference is written as a blank value. After the acquisition is completed, a re-voltage sampling table is generated according to the node number, acquisition time, re-voltage leakage current value, re-voltage partial discharge pulse count, and re-voltage leakage current difference. S4-3. Read the complex voltage sampling table and the lower limit time of complex voltage. First, delete the complex voltage sampling items whose acquisition time is earlier than the lower limit time of complex voltage, and retain the complex voltage sampling items whose acquisition time is later than or equal to the lower limit time of complex voltage. Then, group them by node number, and within the same node, read the retained complex voltage sampling items from earliest to latest acquisition time. Within the same node, the acquisition time of the first complex voltage leakage current difference that is not zero is written as the re-withstand voltage leakage current jump time, and the acquisition time of the first re-withstand voltage partial discharge pulse number that is not zero is written as the re-withstand voltage partial discharge start time. If only the re-withstand voltage leakage current jump time is not obtained, the node will not generate a leakage current advance branch. If only the re-withstand voltage partial discharge start time is not obtained, the node will not generate a partial discharge advance branch. If neither of these is obtained, write the node number, the lower limit time of complex voltage, and the last acquisition time into the complex voltage empty trigger item. S4-4. Read the re-withstand voltage leakage current jump time, re-withstand voltage partial discharge start time, and initial commitment diagram. Search for the same node search node in the initial commitment diagram using the node number. If multiple search nodes exist under the same node number, read the search node with the earliest first abnormal entry time. Read the node number, acquisition time, breakdown state, first abnormal entry time, first leakage current jump time, first partial discharge start time, previous acquisition time summary, and adjacent node summary from the same node search node. Combine these with the field order when generating the node commitment summary in S2 to form the recalculated node input string. Perform a hash digest operation on the recalculated node input string to generate the recalculated node summary. If the recalculated node summary is the same as the node commitment summary in the same node search node, extract the first leakage current jump time and the first partial discharge start time, and combine them with the re-withstand voltage leakage current jump time and the re-withstand voltage partial discharge start time of the same node to form the first voltage recalculated time pair. If the recalculated node summaries are different, or if there is no same node search node in the initial commitment diagram, write the node number into the backtracking mismatch item. S4-5. Read the initial voltage and revoltage timing pair, the lower limit time of revoltage, the recovery lag segment, and the last state signature. Establish a branch certificate for each node. The branches certificate fields include node number, branch type, lower limit time of revoltage, re-voltage withstand voltage anomaly time, first anomaly time, and last state signature. Specifically, for the leakage current advance branch, the re-voltage withstand voltage anomaly time is the re-voltage withstand voltage leakage current jump time, and the first anomaly time is the first leakage current jump time; for the partial discharge advance branch, the re-voltage withstand voltage anomaly time is the re-voltage withstand voltage partial discharge start time, and the first anomaly time is the first partial discharge start time. The re-voltage anomaly time of the absorption current tailing branch is written as null, and the first anomaly time is written as null; when the re-voltage leakage current jump time is later than the lower limit of the re-voltage but earlier than the first leakage current jump time, the corresponding branch certificate is written to the leakage current advance branch; when the re-voltage partial discharge start time is later than the lower limit of the re-voltage but earlier than the first partial discharge start time, the corresponding branch certificate is written to the partial discharge advance branch; when the end of the recovery lag reaches the end of the recovery interval, the corresponding branch certificate is written to the absorption current tailing branch; nodes that do not satisfy any of the above relationships are written to the non-advance branch. Through the above processing, the re-voltage withstand is not a simple repetition of the initial voltage withstand, but rather enters the re-voltage sampling only after the recovery signature chain has been recalculated. The re-voltage withstand sampling item is truncated at the end of the recovery lag segment, so that the extraction of re-voltage withstand anomalies avoids the period when the discharge recovery is not completed. The initial commitment graph provides the initial leakage current jump time and the initial partial discharge start time for each node. The re-voltage withstand sampling table provides the re-voltage withstand leakage current jump time and the re-voltage withstand partial discharge start time for the same node. The dual-anchor certificate shunting algorithm then uses the re-voltage lower bound time and the initial anomaly time together as the branch boundary, thereby distinguishing between leakage current advance, partial discharge advance, absorption current tailing and non-advance. The processing result provides branch certificates and branch types for the branch delimitation and backtracking pruning of S5. In practical applications: if the recovery lag period of a certain connector node ends at 31s, then 31s is written as the lower limit of the overvoltage test. In the withstand voltage sampling, sampling terms earlier than 31s do not participate in branch generation. If the overvoltage leakage current difference of the connector node is not zero at 36s after 31s, and the first leakage current jump time of the same connector node in the initial commitment diagram is 42s, then 36s is later than 31s and earlier than 42s, and the connector node is written into the leakage current advance branch. If another terminal node does not have a leakage current jump and partial discharge start after 31s, then it is written into the overvoltage empty trigger term or non-advance branch.

[0021] S5. Perform branch pruning by recalculating branch certificates, comparing similar time points, and backtracking the initial commitment diagram. Delete branches with broken signatures, out-of-bounds timing, and source mismatch, and output the remaining branches as records of re-energized sensitive insulation weaknesses. In this embodiment, S5 is used to perform certificate recalculation, timing pruning, and commitment backtracking on the partial discharge advance branch, leakage current advance branch, and absorption current tail branch generated by S4. Only branches that pass certificate field recalculation, comparison of similar times, and backtracking of the first commitment map source are written into the re-energized sensitive insulation weakness record. The lower bound certificate read by S5 is the set of certificate fields of the branch certificate in S4. The fields include node number, branch type, lower bound time of re-voltage, time of re-voltage anomaly, time of first similar time, and last status signature. Among them, the branch type is a partial discharge advance branch, leakage current advance branch, or absorption current tail branch. This implementation process includes the following steps: S5-1. After reading the partial discharge advance branch, leakage current advance branch, absorption current tailing branch, and lower bound certificate, extract the certificate field according to the branch type; for the partial discharge advance branch, take the start time of the partial discharge under withstand voltage at the withstand voltage abnormality time, and take the start time of the first similar time; for the leakage current advance branch, take the leakage current jump time under withstand voltage at the withstand voltage abnormality time, and take the first leakage current jump time of the first similar time; for the absorption current tailing branch, write the withstand voltage abnormality time and the first similar time as null values, and read the end time of the recovery lag segment and the last time of the recovery interval; the certificate delimitation recalculation rules are based on node number, branch type, The recalculation certificate is generated in the order of the fields of the lower limit time of the overvoltage test, the re-voltage withstand anomaly time, the first similar time, and the last state signature. If the recalculation certificate is the same as the lower limit certificate, and the re-voltage withstand anomaly time in the partial discharge advance branch or leakage current advance branch is after the lower limit time of the overvoltage test, the delimited branch is output. If the recalculation certificate is different from the lower limit certificate, or the re-voltage withstand anomaly time is not after the lower limit time of the overvoltage test, the branch is deleted, and the node number, branch type, and certificate recalculation failure are written into the signature break item. For the absorption current tailing branch, after the certificate recalculation is passed, it directly enters S5-2 without performing the comparison between the re-voltage withstand anomaly time and the lower limit time of the overvoltage test. S5-2. After reading the delimited branches, execute the same-time cross-pruning rule according to the branch type. Specifically, the re-voltage withstand voltage same-time refers to the re-voltage withstand voltage abnormality time of the branch to be read; the partial discharge advance branch takes the re-voltage withstand voltage partial discharge start time; the leakage current advance branch takes the re-voltage withstand voltage leakage current jump time; the first same-time refers to the first withstand voltage abnormality time of the branch to be read; the partial discharge advance branch takes the first partial discharge start time; the leakage current advance branch takes the first leakage current jump time; the absorption current tailing branch does not take the re-voltage withstand voltage same-time and the first same-time, but compares the end of the recovery lag segment with the last moment of the recovery interval; in the partial discharge advance branch, compare the re-voltage withstand voltage partial discharge start time with the first partial discharge start time; if the re-voltage withstand voltage partial discharge start time is earlier than the first partial discharge start time, output the branch to be traced back; otherwise, delete it. In addition to this branch, the node number, branch type, and partial discharge timing out-of-bounds are written into the timing out-of-bounds item; in the leakage current advance branch, the re-voltage withstand leakage current jump time is compared with the first leakage current jump time. If the re-voltage withstand leakage current jump time is earlier than the first leakage current jump time, the branch to be traced is output; otherwise, the branch is deleted, and the node number, branch type, and leakage current timing out-of-bounds are written into the timing out-of-bounds item; in the absorption current tailing branch, the recovery lag segment endpoint is compared with the last moment of the recovery interval. If the recovery lag segment endpoint reaches the last moment of the recovery interval, the branch to be traced is output; otherwise, the branch is deleted, and the node number, branch type, and tailing not established are written into the timing out-of-bounds item; through this current splitting process, the three types of branches are retained or deleted according to their own fields, and are not replaced by the timing of another branch; S5-3. After reading the branch to be backtracked and the initial commitment graph, execute the commitment counterexample refeedback pruning rules; first, search for the same node search node in the initial commitment graph using the node number in the branch to be backtracked; if there are multiple search nodes with the same node number, take the search node with the earliest initial abnormal entry time; then read the node number, acquisition time, breakdown state, initial abnormal entry time, initial leakage current jump time, initial partial discharge start time, previous acquisition time summary, and adjacent node summary from the same node search node, and form a recalculation input string according to the field order when generating the node commitment summary in S2; perform hash digest operation on the recalculation input string to obtain the recalculated node commitment summary; the recalculated node commitment summary... If the node commitment summary is the same as that of the node in the same node search node, the node number, node type, start position, end position, branch type, first similar time, re-voltage withstand anomaly time, re-voltage lower limit time, recovery lag segment, and lower limit certificate are written into the re-energization sensitive insulation weakness record; if the recalculated node commitment summary is different, or if there is no same node search node in the first commitment diagram, the branch is deleted, and the node number, branch type, and backtracking mismatch are written as the deletion reason into the parent search node to form a counterexample summary; the parent search node is the search node in the first commitment diagram that is in the upper-level merging relationship with the node number; if there is no upper-level merging relationship, the root node corresponding to the first commitment root signature is taken; Through the above processing, S5 sequentially performs certificate recalculation, same-time pruning, and first commitment graph backtracking on the three types of branches generated by S4, so that the retained branches not only satisfy the condition that they occur after the lower bound of the recalculation, but also satisfy the condition that they occur earlier than the first same-time, and can be backtracked to the same node search node in the first commitment graph; the deleted branches are not directly discarded, but the deletion reason is recorded as a signature break item, time sequence out-of-bounds item, or counterexample summary, so that subsequent same-source branches can reduce repeated backtracking along the parent search node; In practical applications: In the leakage current advance branch of a certain joint node, the lower limit time of the re-voltage is 31s, the leakage current jump time of the re-voltage withstand is 36s, and the first leakage current jump time is 42s. After the certificate recalculation, it is the same as the lower limit certificate. Then the branch is first delimited by the certificate. If 36s is earlier than 42s, it is then pruned by cross-pruning at the same time. Then, the first commitment diagram is traced back with the joint node number. If the recalculated node commitment summary is the same as the node commitment summary in the first commitment diagram, then the joint node is written into the re-energized sensitive insulation weakness record, and the branch type is written as leakage current advance branch.

[0022] Furthermore, a withstand voltage testing system for insulated power cables includes: The node sequencing module is used to acquire the object record, withstand voltage, holding time, discharge conditions and recovery interval of the cable under test, split the object record into a node sequence according to the body, terminal and connector, and generate an initial signature; The first-voltage verification module is used to perform the first withstand voltage test according to the withstand voltage and holding time, collect the breakdown state, leakage current jump time and partial discharge start time of each node, and use the Merkle-DAG node commitment pruning rules to write the node number, acquisition time and acquisition results into the search node to generate the first commitment graph. The discharge verification module is used to perform discharge recovery according to the discharge conditions when the initial withstand voltage does not break down. It collects the residual voltage decay sequence and the absorption current fallback sequence, and uses the hash chain state signature rule to sequentially write the initial signature, the initial commitment graph digest and the discharge recovery data into the hash chain to generate the recovery signature chain. The re-voltage shunt module is used to perform re-voltage withstand based on the recovery signature chain, so that the tested cable is brought back to the same withstand voltage. It collects the re-voltage withstand leakage current jump time and the re-voltage withstand partial discharge start time, and generates a branch certificate by signing the node number, branch type, re-voltage lower limit time, re-voltage abnormal time, first abnormal time and last status signature. It also generates partial discharge early branch, leakage current early branch and absorption current tailing branch. The backtracking pruning module is used to backtrack and perform branch pruning based on branch certificate recalculation, comparison of similar time periods, and initial commitment graph. It deletes branches with broken signatures, out-of-bounds timing, and source mismatch, and outputs the remaining branches as re-energized sensitive insulation weakness records.

[0023] Working principle: This scheme first disassembles the cross-linked polyethylene insulated power cable and cable accessories into a node sequence according to the body, terminals and joints, and writes the withstand voltage, holding time, discharge conditions and recovery interval into the initial signature; then, the first withstand voltage is performed, the breakdown state, leakage current jump time and partial discharge start time of each node are collected, and the first commitment map is generated by searching the nodes through Merkle-DAG, which serves as the first voltage benchmark for subsequent withstand voltage comparison; if the first withstand voltage does not break down, discharge recovery is performed, and the recovery lag segment is truncated according to the difference between the residual voltage decay and the absorption current fall, and a recovery signature chain is generated; then, the withstand voltage is increased to the same withstand voltage and the re-withstand voltage is performed, only the re-voltage sampling results after the recovery lag segment are read, it is determined whether the leakage current jump or partial discharge start in the re-withstand voltage is earlier than the first similar time, and invalid branches are eliminated by branch certificate, timing pruning and commitment backtracking, and finally the record of the insulation weakness of the re-energized sensitive type is output; In practical applications, after the joint fabrication and terminal installation of a newly laid cross-linked polyethylene insulated power cable and its accessories are completed, the conventional withstand voltage test may show that the cable did not break down during the first test. However, there may be a tiny interface gap inside a certain intermediate joint, which may not develop into a breakdown during the first continuous test. When using this solution, the system first records the leakage current and partial discharge initiation time of the joint during the first withstand voltage test, and then observes the recovery of the residual voltage and absorbed current after discharge. If there is a lag in the discharge recovery, the system uses the lag endpoint as the effective starting point for reading the re-withstand voltage test. After the withstand voltage is increased again, if the leakage current jump or partial discharge initiation of the joint occurs earlier than the first withstand voltage test, and the result can be traced back to the same joint node in the initial commitment diagram, then the joint is recorded as a re-energized sensitive insulation weakness, which is used to indicate that it needs to be carefully reviewed before decommissioning or commissioning.

[0024] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for withstand voltage testing of insulated power cables, characterized in that, include: S1. Obtain the object record, withstand voltage, holding time, discharge conditions and recovery interval of the cable under test, split the object record into node sequences according to the body, terminal and connector, and generate an initial signature; S2. Perform the first withstand voltage test according to the withstand voltage and holding time, collect the breakdown state, leakage current jump time and partial discharge start time of each node, and use the Merkle-DAG node commitment pruning rules to write the node number, collection time and collection results into the search node to generate the first commitment graph. S3. If the initial withstand voltage test fails to break down, perform discharge recovery according to the discharge conditions, collect the residual voltage value and the absorption current value, extract the recovery hysteresis segment according to the residual voltage drop and the absorption current drop, and write the initial signature, the initial commitment diagram and the recovery hysteresis segment into the status item to generate the recovery signature chain. S4. Based on the recovery signature chain, perform re-withstand voltage to bring the tested cable back to the same withstand voltage. Collect the re-withstand voltage leakage current jump time and the re-withstand voltage partial discharge start time. Generate a branch certificate by signing the node number, branch type, re-voltage lower limit time, re-withstand voltage abnormal time, first abnormal time and last status. Generate partial discharge early branch, leakage current early branch and absorption current tailing branch. S5. Perform branch pruning by recalculating branch certificates, comparing similar time points, and backtracking the initial commitment graph. Delete branches with broken signatures, out-of-bounds timing, and source mismatch, and output the remaining branches as re-energized sensitive insulation weakness records.

2. The withstand voltage test method for an insulated power cable according to claim 1, characterized in that: S1 includes: S1-1. Read the starting terminal position, ending terminal position, connector position and body length from the object record. Sort the starting terminal position, connector position and ending terminal position in ascending order of distance along the line. Bind the withstand voltage, holding time, discharge condition and recovery interval to the sorting result and output the position sequence. S1-2. Read the position sequence, write the starting and ending positions as terminal nodes, write the joint position as joint nodes, write the line segment between two adjacent positions as body nodes, and write the node number, node type, starting position, ending position and adjacent node number for each node, and output the node sequence. S1-3. Read the node sequence, withstand voltage, holding time, discharge condition and recovery interval. Generate the first node digest by hash digest operation using withstand voltage, holding time, discharge condition, recovery interval and the first node field. Then, starting from the second node, recursively generate the current node digest by hash digest operation using the previous node digest and the current node field. Output the last node digest as the initial signature.

3. The withstand voltage test method for an insulated power cable according to claim 2, characterized in that: S2 includes: S2-1. Read the node sequence, withstand voltage and holding time to perform the first withstand voltage test. Write the breakdown status, leakage current value and partial discharge pulse number according to the node number and the acquisition time to generate the first voltage sampling table. S2-2. Read the first voltage sampling table, calculate the difference between the current leakage current value and the previous leakage current value in ascending order of the sampling time of the same node, write the first sampling time when the difference is not zero as the leakage current jump time, write the first sampling time when the number of partial discharge pulses is not zero as the partial discharge start time, and generate the first abnormal entry time according to the earlier of the two times.

4. The withstand voltage test method for an insulated power cable according to claim 3, characterized in that: S2 further includes: S2-3. Read the first abnormal entry time, breakdown status and initial signature, write the node number, first abnormal entry time, leakage current jump time, partial discharge start time, previous acquisition time summary and adjacent node summary into the Merkle-DAG search node, and generate the node commitment summary through hash digest operation; S2-4. Read the Merkle-DAG search nodes. Group the search nodes with the same node number, breakdown state, leakage current jump time and partial discharge start time, and whose node commitment summary is derived from the summary of the same previous acquisition time and the summary of the same adjacent node into the same certificate node group. Retain the search node in the same certificate node group whose first abnormal entry time is earlier. Connect the retained search nodes according to the node sequence to generate the first commitment map as the benchmark for early re-withstand voltage determination.

5. The withstand voltage test method for an insulated power cable according to claim 4, characterized in that: S3 includes: S3-1. When the breakdown state in the initial commitment diagram is non-breakdown, read the discharge condition and recovery interval from the object record, break down the discharge condition into residual voltage limit and sampling period, and merge the node commitment summary of each search node in the initial commitment diagram level by level according to the node sequence to generate the initial commitment root signature and discharge task. S3-2. Perform discharge recovery according to the discharge task. Write the sampling time, residual voltage value and absorption current value according to the sampling period. Subtract the current residual voltage value from the previous residual voltage value to generate the residual voltage drop value. Subtract the current absorption current value from the previous absorption current value to generate the absorption current drop value. Output the recovery sampling table.

6. The withstand voltage test method for an insulated power cable according to claim 5, characterized in that: S3 further includes: S3-3. Read the recovery sampling table, write the first sampling time when the residual voltage value is not greater than the residual voltage limit as the discharge end point, calculate the difference between the residual voltage drop and the current absorption drop from the discharge end point according to the sampling time, write the sampling time when the difference changes from positive to negative as the recovery lag start point, write the sampling time when the difference changes from negative to positive as the recovery lag end point, and output the recovery lag segment. S3-4. Write the initial signature, the first commitment root signature, the recovery lag start point, the recovery lag end point, and the residual voltage drop and current drop within the recovery lag segment into the recovery status item according to the acquisition time. The current recovery status item is generated by the previous recovery signature and the current recovery status item. The recovery lag end point is output as the lower bound time of the re-withstand voltage advance, and the recovery signature chain is generated.

7. The withstand voltage test method for an insulated power cable according to claim 6, characterized in that: S4 includes: S4-1. Read the recovery lag segment, recovery interval, state item and last state signature from the recovery signature chain. Generate a recalculation signature by taking the item number, previous signature and recovery field in each state item in a fixed field order. When the recalculation signature is the same as the state signature in the same order in the recovery signature chain, write the end point of the recovery lag segment as the lower bound of the recalculation and generate a recalculation token. Otherwise, generate a recalculation blocking item. S4-2. After generating the re-voltage token, perform re-voltage withstand to bring the tested cable back to the same withstand voltage. Write the re-voltage leakage current value and the re-voltage partial discharge pulse number according to the acquisition time. Subtract the previous re-voltage leakage current value from the current re-voltage leakage current value to generate the re-voltage leakage current difference value, and output the re-voltage sampling table. S4-3. Read the complex voltage sampling table and the lower limit time of complex voltage. Use the lower limit window timing algorithm to delete complex voltage sampling items whose acquisition time is earlier than the lower limit time of complex voltage. Among the remaining complex voltage sampling items, take the first acquisition time when the complex voltage leakage current difference is not zero as the re-voltage withstand leakage current jump time. Take the first acquisition time when the re-voltage withstand partial discharge pulse number is not zero as the re-voltage withstand partial discharge start time. If no time is obtained, generate a complex voltage empty trigger item.

8. The withstand voltage test method for an insulated power cable according to claim 7, characterized in that: S4 further includes: S4-4. Read the re-voltage withstand leakage current jump time, the re-voltage withstand partial discharge start time, and the first commitment diagram. Use the DAG anchor point backtracking algorithm to search for the same node in the first commitment diagram by node number. Generate a recalculated node summary by combining the node number, the first leakage current jump time, the first partial discharge start time, the previous acquisition time summary, and the adjacent node summary in the same node search node. If the recalculated node summary is the same as the node commitment summary, generate the first voltage re-voltage time pair; otherwise, generate a backtracking mismatch item. S4-5. Read the initial voltage and re-voltage timing pair, the lower limit of the re-voltage, the recovery lag segment, and the last state signature. Use the dual-anchor certificate splitting algorithm to generate a branch certificate by combining the node number, branch type, lower limit of the re-voltage, re-voltage anomaly timing, first anomaly timing, and last state signature. When the re-voltage leakage current jump timing is later than the lower limit of the re-voltage but earlier than the first leakage current jump timing, write the branch certificate to the leakage current advance branch. When the re-voltage partial discharge start timing is later than the lower limit of the re-voltage but earlier than the first partial discharge start timing, write the branch certificate to the partial discharge advance branch. When the end of the recovery lag segment reaches the last moment of the recovery interval, write the branch certificate to the absorption current tailing branch. Write the remaining nodes to the non-advance branch.

9. The withstand voltage test method for an insulated power cable according to claim 8, characterized in that: S5 includes: S5-1: Read the partial discharge advance branch, leakage current advance branch, absorption current tailing branch and lower limit certificate. Generate a recalculation certificate by taking the node number, branch type, lower limit time of the re-voltage withstand voltage abnormality time, first similar time and last status signature in a fixed field order. If the recalculation certificate is the same as the lower limit certificate and the re-voltage withstand voltage abnormality time is after the lower limit time of the re-voltage withstand voltage, output the delimited branch. Otherwise, delete the branch and output the signature break item. S5-2. Read the bounded branches and adopt the same-time cross-pruning rule. Compare the start time of the re-withstand voltage partial discharge in the partial discharge advance branch with the start time of the first partial discharge. Compare the re-withstand voltage leakage current jump time in the leakage current advance branch with the first leakage current jump time. Compare the end time of the recovery lag segment in the absorption current tailing branch with the end time of the recovery interval. If the re-withstand voltage same-time is earlier than the first same-time or the end time of the recovery lag segment reaches the end time of the recovery interval, output the branch to be traced back. Otherwise, delete the branch and output the timing out-of-bounds item. S5-3. Read the branch to be traced back and the initial commitment graph. Using the commitment counterexample refeedback pruning rule, trace the node number of the branch to be traced back to the same node search node in the initial commitment graph. Then, recalculate the node commitment summary using the node number of the same node search node, the time of the first leakage current jump, the time of the first partial discharge start, the summary of the previous acquisition time, and the summary of adjacent nodes. If the recalculated node commitment summary is the same as the node commitment summary of the same node search node, output the re-energized sensitive insulation weakness record. Otherwise, delete the branch and write the node number, branch type, and deletion reason into the parent search node to form a counterexample summary.

10. A withstand voltage testing system for insulated power cables, characterized in that, include: The node sequencing module is used to acquire the object record, withstand voltage, holding time, discharge conditions and recovery interval of the cable under test, split the object record into a node sequence according to the body, terminal and connector, and generate an initial signature; The first-voltage verification module is used to perform the first withstand voltage test according to the withstand voltage and holding time, collect the breakdown state, leakage current jump time and partial discharge start time of each node, and use the Merkle-DAG node commitment pruning rules to write the node number, acquisition time and acquisition results into the search node to generate the first commitment graph. The discharge verification module is used to perform discharge recovery according to the discharge conditions when the initial withstand voltage does not break down. It collects the residual voltage decay sequence and the absorption current fallback sequence, and uses the hash chain state signature rule to sequentially write the initial signature, the initial commitment graph digest and the discharge recovery data into the hash chain to generate the recovery signature chain. The re-voltage shunt module is used to perform re-voltage withstand based on the recovery signature chain, so that the tested cable is brought back to the same withstand voltage. It collects the re-voltage withstand leakage current jump time and the re-voltage withstand partial discharge start time, and generates a branch certificate by signing the node number, branch type, re-voltage lower limit time, re-voltage abnormal time, first abnormal time and last status signature. It also generates partial discharge early branch, leakage current early branch and absorption current tailing branch. The backtracking pruning module is used to backtrack and perform branch pruning based on branch certificate recalculation, comparison of similar time periods, and initial commitment graph. It deletes branches with broken signatures, out-of-bounds timing, and source mismatch, and outputs the remaining branches as re-energized sensitive insulation weakness records.

Images