Medium and small river hydrological situation change analysis method based on improved hydrological variation index system

Abstract

The invention discloses a medium and small river hydrological situation change analysis method based on an improved hydrological variation index system, and the method comprises the steps: firstly, carrying out hydrological staging by taking ten days as a calculation scale according to the seasonal characteristics of medium and small river runoff, and dividing a year into a dry period, a previous flood period, a main flood period and a later flood period; then, hydrological parameters in each stage are counted, parameter supplementation is carried out on a conventional hydrological variation index system, and an improved hydrological variation index system is provided; and finally, based on the improved hydrological variation index system, calculating the change degree of the hydrological conditions of the medium and small rivers by adopting a change range evaluation method and a histogram matching method. According to the improved hydrological variation index system provided by the invention, hydrological situation characteristics of medium and small rivers can be comprehensively represented, and the situation that seasonal differences of parameters in different statistical periods are ignored due to the fact that a conventional hydrological variation index system adopts a single statistical scale is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrological evaluation, and in particular to a method for evaluating the degree of change of hydrological conditions in small and medium-sized rivers. Background Art

[0002] The hydrological regime is the main driving force of the river ecosystem structure and is crucial to maintaining river health and ecosystem functions. However, water conservancy projects built by damming rivers block the information and material exchange between upstream and downstream rivers, significantly changing the hydrological regime of rivers, and thus causing a series of river ecological and environmental problems. Carrying out analysis of hydrological regime changes and evaluating the impact of water conservancy projects on river ecosystems are of great significance for formulating river ecological protection measures. At present, the analysis and research of hydrological regime changes are mainly focused on large rivers, and the research on small and medium-sized rivers is relatively insufficient. Due to the small catchment area and fast confluence speed, the runoff process of small and medium-sized rivers is greatly affected by rainfall factors, and there is a multi-peak phenomenon, especially during the flood season when extreme rainfall is frequent, the multi-peak phenomenon of runoff is more prominent. The conventional hydrological variation index system represents the hydrological information comprehensively, scientifically, and easily used, and is widely used in river hydrological regime analysis. However, the method of statistically analyzing each parameter in units of years will ignore the seasonal differences of some hydrological parameters, and it is difficult to accurately reflect the hydrological regime characteristics of small and medium-sized rivers. Therefore, it is necessary to propose a hydrological variation index system suitable for small and medium-sized rivers, comprehensively consider the seasonal differences of various hydrological parameters, and comprehensively reflect the hydrological regime characteristics of rivers. Summary of the invention

[0003] Purpose of the invention: To provide a hydrological situation analysis method based on an improved hydrological variation index system, and to provide a feasible method and technology for accurately evaluating the changing laws of hydrological situations in small and medium-sized rivers.

[0004] Technical solution: A hydrological situation analysis method based on an improved hydrological variation index system of the present invention comprises the following steps:

[0005] S1 collects historical runoff data and constructs a hydrological stage index set X. The CRITIC method is used to calculate the weight of each index, and the stage index set X is converted into a stage comprehensive index set Y. The threshold L for judging the flood season and the dry season is constructed from Y, and the membership of each decade to the flood season set is calculated.

[0006] Furthermore, the hydrological stage indicator set X constructed in step S1 is as follows:

[0007]

[0008] Where p is the number of years of runoff data; X iis the indicator set for the i-th year; m is the number of indicators selected every ten days (based on the average flow rate, the minimum and maximum 1-day, 3-day, 5-day, 7-day runoff and other indicators that can reflect the runoff process information are added); It is the j-th indicator in the k-th decade of the i-th year.

[0009] Furthermore, the steps of converting the hydrological stage indicator set X into the stage comprehensive indicator set Y in step S1 are as follows:

[0010]

[0011] Among them, Y i is the comprehensive indicator set for the i-th year; is the comprehensive index of the kth decade in the i-th year; W j is the weight of the j-th indicator, calculated using the CRITIC method.

[0012] Furthermore, the calculation formula of the threshold L used to judge the flood season and the dry season in step S1 is as follows:

[0013]

[0014] in, is the mean value of the jth indicator.

[0015] Furthermore, the steps of calculating the membership of each decade to the flood season set in step S1 are as follows:

[0016] ① For a given runoff sequence in year i, select a comprehensive indicator: The starting period is greater than the threshold L and end of the decade As the beginning and end of the flood season, and so on, the series data of p years can be statistically obtained into p flood season intervals Where i=1,2,...,p.

[0017] ② Count the number n of times each decade falls within the p flood season intervals as its degree of membership to the flood season.

[0018] λ k =n / p (4)

[0019] S2 is based on the membership sequence {λ k}, divided into dry season and flood season. And further divided the flood season into pre-flood season, main flood season and post-flood season by using mean change point analysis method.

[0020] Furthermore, the criteria for dividing into dry season and flood season in step S2 are as follows:

[0021] Select the membership degree λ k= 0.5 as the cutoff level to determine whether it is a flood season or not. k >0.5, it means that the kth decade has a stronger tendency towards the flood season, and the kth decade is included in the flood season; when λ k When <0.5, it means that the kth decade has a stronger tendency towards dry period, and the kth decade is classified as dry period.

[0022] Furthermore, the steps of using the mean change point analysis method to divide the flood season into the pre-flood season, the main flood season and the post-flood season in step S2 are as follows:

[0023] ① Take the ten-day period with the largest membership degree to the flood season m As the demarcation point, the membership sequence {λ k} is divided into the pre-sequence {λ f ,λ f+1 ,...,λ m} and the subsequent sequence {λ m ,λ m+1 ,...,λ f+d}. Among them, f and d represent the beginning and length of the flood season.

[0024] ② Assume that the previous sequence {λ f ,λ f+1 ,...,λ m The change point of} is the αth decadeλ α , calculate the difference β between the statistics D and D'.

[0025]

[0026] β=DD' (6)

[0027] in, are the sequences {λ f ,λ f+1 ,...,λ α}、{λ α+1 ,λ α+1 ,...,λ m} and {λ f ,λ f+1 ,...,λ m}.

[0028] ③ The change point α changes from f to m in sequence, and the decade s corresponding to the maximum β is selected as the dividing point between the pre-flood and the main flood.

[0029] ④ Similarly, according to the above change point search method, obtain the posterior sequence {λ m ,λ m+1 ,...,λ f+d}, and finally complete the division of the flood season. The pre-flood season, main flood season and post-flood season correspond to [f,s], [s+1,e] and [e+1,f+d] respectively.

[0030] Based on the conventional hydrological variation index system, S3 keeps the parameters of the first and third components unchanged, supplements the parameters of the second, fourth and fifth components, and proposes an improved hydrological variation index system.

[0031] Furthermore, the first group of indicator systems reflects the basic hydrological situation requirements of the river from the perspective of the average monthly flow. Since it clarifies the scale of the parameters, it does not change and includes the average monthly flow of 12 months. The third group of indicator systems reflects the extreme hydrological situation requirements of the river, conveying the migration and spawning signals of organisms. Usually, biological migration and spawning are not frequent and are only related to the extreme flow within the year. Therefore, the statistical scale of the parameters is still mainly based on the year, including two parameters: the time when the annual maximum and annual minimum flow occur. The second group adds the maximum and minimum 1-day, 3-day, and 7-day flows in the four stages of dry season, pre-flood season, main flood season, and post-flood season, adding a total of 24 parameters. The fifth group adds the average flow increase rate, average flow decrease rate, and flow reversal times in the four stages of dry season, pre-flood season, main flood season, and post-flood season, adding a total of 12 parameters. The fourth group reflects the high and low flow pulse process, and the short statistical scale is easy to cause the fragmentation of flow pulse events. Therefore, the flood season is no longer divided into three stages: pre-flood, main flood and post-flood. Instead, only the number and duration of high and low flow pulses in the dry season and flood season are increased, with a total of 8 parameters added.

[0032] S4 is based on an improved hydrological variation index system and uses the RVA method (i.e., range of variation assessment method, the same below) and the HMA method (i.e., histogram matching method, the same below) to calculate the degree of change in the hydrological situation of small and medium-sized rivers.

[0033] Furthermore, in step S4, based on the improved hydrological variation index system, the steps of using the RVA method to calculate the degree of change in the hydrological situation of small and medium-sized rivers are as follows:

[0034] ① Determine the mutation point of the runoff sequence and calculate the parameter values ​​of each component under the improved hydrological variation index system based on the historical daily flow data before the runoff change.

[0035] ② Select the characteristic values ​​of each parameter at 25% and 75% frequency to form the target range I of RVA, which is the acceptable range of each hydrological parameter.

[0036] ③ Based on the changed historical daily flow data, calculate the parameter values ​​of each component under the improved hydrological variation index system after the runoff is affected, and calculate the hydrological change degree of a single parameter and the overall hydrological change degree according to the following formula.

[0037]

[0038] Among them, D iis the hydrological change degree of the i-th parameter; D is the overall hydrological change degree; m is the number of parameters in the improved hydrological variation index system; N io is the number of years in which the i-th parameter actually falls within the RVA target range I; N ie The number of years in which the ith parameter is expected to fall within the RVA target range can be taken as half of the total number of years after the impact. The classification criteria for the overall hydrological change of the river include slight change (0-33%), moderate change (33%-67%) and high change (67%-100%).

[0039] Furthermore, in step S4, based on the improved hydrological variation index system, the steps of using the HMA method to calculate the degree of change in the hydrological regime of small and medium-sized rivers are as follows:

[0040] ①Determine the number of classes n in the histogram c , that is, how many intervals the parameter should be divided into.

[0041]

[0042] Among them, r is the difference between the maximum and minimum values; n is the number of data; r iq It is the difference between the 3 / 4 quantile and the 1 / 4 quantile.

[0043] ② Draw the frequency histograms H and K of the hydrological parameters before and after the change, and use the quadratic distance d Q (H,K) measures the difference between two histograms.

[0044]

[0045] Where h=(h1,h2,...,h nc )、k=(k1,k2,...,k nc ) are the histogram frequency vectors of the parameters before and after the change; A=[a ij ] is the similarity matrix, a ij is the similarity between the i-th category and the j-th category in the histogram, calculated as follows:

[0046]

[0047] Among them, d ij is the distance between the i-th class and the j-th class in the histogram.

[0048] ③ Calculate the hydrological change degree D of each parameter Q,p .

[0049]

[0050] Among them, d Q,p is the difference of the frequency histogram before and after the pth hydrological parameter changes.

[0051] ④Calculate the overall hydrological change.

[0052]

[0053] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0054] (1) Traditional staging methods often use the average flow rate as a staging index. A single staging index does not fully consider the runoff process within a ten-day period, and ignores information such as extreme flows and high and low flow pulses within a ten-day period. The present invention uses the ten-day period as the calculation scale. On the basis of the average flow rate, it adds indicators such as the minimum and maximum 1-day, 3-day, 5-day, and 7-day runoffs that can reflect the runoff process information. The constructed hydrological staging index set has an appropriate scale, reasonable indicators, and more accurate staging results.

[0055] (2) The conventional hydrological variation index system uses the method of statistically analyzing each parameter in years, which will ignore the seasonal differences of some hydrological parameters and make it difficult to accurately reflect the hydrological characteristics of small and medium-sized rivers. The present invention supplements and improves the conventional hydrological variation index system by statistically analyzing the hydrological parameters in each period. The proposed improved hydrological variation index system can comprehensively reflect the hydrological characteristics of rivers. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of the method of the present invention;

[0057] Figure 2 It is a diagram of the hydrological stages of Liujiaping River in an embodiment of the present invention;

[0058] Figure 3 The degree of change of some hydrological parameters of Liujiaping River before and after the reservoir is built under the original statistical scale (full year) and the newly added statistical scale (dry season, pre-flood season, main flood season, post-flood season) in the embodiment of the present invention;

[0059] Figure 4 In the embodiment of the present invention, based on the conventional and improved hydrological variation index systems, the RVA method and the HMA method are used to quantitatively calculate the degree of change of the overall hydrological situation of the Liujiaping River before and after the reservoir is built. DETAILED DESCRIPTION

[0060] The technical solution of the present invention is described in detail below, but the protection scope of the present invention is not limited to the embodiments.

[0061] like Figure 1 As shown, a method for analyzing changes in hydrological conditions of small and medium-sized rivers based on an improved hydrological variation index system of the present invention comprises the following steps:

[0062] S1 collects historical runoff data and constructs a hydrological stage index set X. The CRITIC method is used to calculate the weight of each index, and the stage index set X is converted into a stage comprehensive index set Y. The threshold L for judging the flood season and the dry season is constructed from Y, and the membership of each decade to the flood season set is calculated.

[0063] The constructed hydrological stage indicator set X is as follows:

[0064]

[0065] Where p is the number of years of runoff data; X i is the indicator set for the i-th year; m is the number of indicators selected every ten days (based on the average flow rate, the minimum / maximum 1-day, 3-day, 5-day, 7-day runoff and other indicators that can reflect the runoff process information are added); It is the j-th indicator in the k-th decade of the i-th year.

[0066] The steps to transform the hydrological stage indicator set X into the stage comprehensive indicator set Y are as follows:

[0067]

[0068] Among them, Y i is the comprehensive indicator set for the i-th year; is the comprehensive index of the kth decade in the i-th year; W j is the weight of the j-th indicator, calculated using the CRITIC method.

[0069] The calculation formula of the threshold L used to judge the flood season and dry season is as follows:

[0070]

[0071] in, is the mean value of the jth indicator.

[0072] The calculation steps of the membership of each decade to the flood season set are as follows:

[0073] ① For a given runoff sequence in year i, select a comprehensive indicator: The starting period is greater than the threshold L and end of the decade As the beginning and end of the flood season, and so on, the series data of p years can be statistically obtained into p flood season intervals Where i=1,2,...,p.

[0074] ② Count the number n of times each decade falls within the p flood season intervals as its degree of membership to the flood season.

[0075] λ k =n / p (4)

[0076] S2 is based on the membership sequence {λ k}, divided into dry season and flood season. And further divided the flood season into pre-flood season, main flood season and post-flood season by using mean change point analysis method.

[0077] Select the membership degree λ k = 0.5 as the cutoff level to determine whether it is a flood season or not. k >0.5, it means that the kth decade has a stronger tendency towards the flood season, and the kth decade is included in the flood season; when λ k When <0.5, it means that the kth decade has a stronger tendency towards dry period, and the kth decade is classified as dry period.

[0078] The steps of dividing the flood season into pre-flood season, main flood season and post-flood season using mean change point analysis are as follows:

[0079] ① Take the ten-day period with the largest membership degree to the flood season m As the demarcation point, the membership sequence {λ k} is divided into the pre-sequence {λ f ,λ f+1 ,...,λ m} and the subsequent sequence {λ m ,λ m+1 ,...,λ f+d}. Among them, f and d represent the beginning and length of the flood season.

[0080] ② Assume that the previous sequence {λ f ,λ f+1 ,...,λ m The change point of} is the αth decadeλ α , calculate the difference β between the statistics D and D'.

[0081]

[0082] β=DD' (6)

[0083] in, are the sequences {λ f ,λ f+1 ,...,λ α}、{λ α+1 ,λ α+1 ,...,λ m} and {λ f ,λ f+1 ,...,λ m}.

[0084] ③ The change point α is from f to m in sequence, and the decade s corresponding to the maximum β is selected as the change point of the previous sequence, which serves as the dividing point between the pre-flood season and the main flood season.

[0085] ④ Similarly, according to the above change point search method, obtain the posterior sequence {λ m ,λ m+1 ,...,λ f+d}, and finally the flood season division is completed. The pre-flood season, main flood season and post-flood season correspond to [f,s], [s+1,e] and [e+1,f+d] respectively.

[0086] Based on the conventional hydrological variation index system, S3 keeps the parameters of the first and third components unchanged, supplements the parameters of the second, fourth and fifth components, and proposes an improved hydrological variation index system.

[0087] The second group adds the maximum and minimum 1-day, 3-day and 7-day flows in the four stages of dry season, pre-flood season, main flood season and post-flood season, adding a total of 24 parameters.

[0088] The fourth component adds the number and duration of high and low flow pulses in the dry season and flood season, adding a total of 8 parameters.

[0089] The fifth group adds the average flow increase rate, average flow decrease rate and flow reversal times in the four stages of dry season, pre-flood season, main flood season and post-flood season, adding a total of 12 parameters.

[0090] S4 is based on an improved hydrological variation index system and uses the RVA method and the HMA method to calculate the degree of change in the hydrological regime of small and medium-sized rivers.

[0091] The steps for calculating the degree of change in the hydrological regime of small and medium-sized rivers using the RVA method are as follows:

[0092] ① Determine the mutation point of the runoff sequence and calculate the parameter values ​​of each component under the improved hydrological variation index system based on the historical daily flow data before the runoff change.

[0093] ② Select the characteristic values ​​of each parameter at 25% and 75% frequency to form the target range I of RVA, which is the acceptable range of each hydrological parameter.

[0094] ③ Based on the changed historical daily flow data, calculate the parameter values ​​of each component under the improved hydrological variation index system after the runoff is affected, and calculate the hydrological change degree of a single parameter and the overall hydrological change degree according to the following formula.

[0095]

[0096] Among them, D i is the hydrological change degree of the i-th parameter; D is the overall hydrological change degree; m is the number of parameters in the improved hydrological variation index system; N io is the number of years in which the i-th parameter actually falls within the RVA target range I; N ieThe number of years in which the ith parameter is expected to fall within the RVA target range can be taken as half of the total number of years after the impact. The classification criteria for the overall hydrological change of the river include slight change (0-33%), moderate change (33%-67%) and high change (67%-100%).

[0097] The steps for calculating the degree of change in the hydrological regime of small and medium-sized rivers using the HMA method are as follows:

[0098] ①Determine the number of classes n in the histogram c , that is, how many intervals the parameter should be divided into.

[0099]

[0100] Among them, r is the difference between the maximum and minimum values; n is the number of data; r iq It is the difference between the 3 / 4 quantile and the 1 / 4 quantile.

[0101] ② Draw the frequency histograms H and K of the hydrological parameters before and after the change, and use the quadratic distance d Q (H,K) measures the difference between two histograms.

[0102]

[0103] Where h=(h1,h2,...,h nc )、k=(k1,k2,...,k nc ) are the histogram frequency vectors of the parameters before and after the change; A=[a ij ] is the similarity matrix, a ij is the similarity between the i-th category and the j-th category in the histogram, calculated as follows:

[0104]

[0105] Among them, d ij is the distance between the i-th class and the j-th class in the histogram.

[0106] ③ Calculate the hydrological change degree D of each parameter Q,p .

[0107]

[0108]

[0109] Among them, d Q,p is the difference of the frequency histogram before and after the pth hydrological parameter changes.

[0110] ④Calculate the overall hydrological change.

[0111]

[0112] Example:

[0113] This embodiment is illustrated by taking the Liujiaping River in Hunan Province as an example. The river is a typical small and medium-sized river. The improved hydrological variation index system proposed in the present invention is used to analyze the changing patterns of the river hydrological situation before and after the reservoir construction, illustrating the effectiveness and rationality of the present invention.

[0114] Liujiaping River is located in Xupu County, Hunan Province, at 110°39'~110°43' east longitude and 27°27'~27°30' north latitude, with a total drainage area of ​​72km 2 The river is 30km long, with a large slope and an average flow of 1.51m over many years. 3 / s. Liujiaping Power Station was built on the river. The construction of the power station started in 1984 and it stored water for power generation in 1987. Reservoir construction and power station operation are important factors affecting the changes in the river's hydrological regime. Therefore, the daily runoff from 1962 to 1981 was taken as the natural runoff, and the daily discharge of the power station from 1999 to 2018 was taken as the changed runoff to analyze the changes in the hydrological regime of Liujiaping River before and after the reservoir construction.

[0115] The daily runoff from 1962 to 1981 was used as sample data, and 720 ten-day samples were generated in units of ten days. Then five indicators, including the ten-day average runoff, the ten-day maximum and minimum one-day runoff, and the ten-day maximum and minimum three-day runoff, were selected to form the period indicator set X. (720×5) The weights of the five indicators of average runoff per decade, maximum and minimum one-day runoff per decade, and maximum and minimum three-day runoff per decade calculated by the CRITIC method are 0.1982, 0.1011, 0.2859, 0.1331, and 0.2816, respectively. The comprehensive index value of each decade sample is calculated according to formula (2), the stage threshold L = 5.503 is calculated according to formula (3), and the membership of each decade to the flood season set is calculated according to formula (4). The results are shown in Table 1.

[0116] Table 1 The membership degree of each decade to the flood season set

[0117]

[0118]

[0119] Figure 2 The calculation process of hydrological period is shown, with the membership degree λ of each decade to the flood season k= 0.5 as the cutoff level for flood season division, the flood season is from late April to mid-August, and the rest of the time period is the dry season. Further, with the maximum membership degree within the flood season as the dividing point, the ordered membership degree sequence of the flood season is divided into the front sequence (late April to early June) and the back sequence (mid-June to mid-August), and according to formula (5) and formula (6), the change points of the front and back sequences are found, which are late May and early July respectively. Therefore, the result of the division within the flood season is: late April to late May belongs to the front flood season, early June to early July belongs to the main flood season, and mid-July to mid-August belongs to the back flood season.

[0120] Taking each hydrological stage as the statistical scale, the parameters of the conventional hydrological variation index system are supplemented. The second component adds the maximum and minimum 1-day, 3-day, and 7-day flows in the four stages of dry season, pre-flood season, main flood season, and post-flood season, totaling 24 parameters; the fourth component adds the number of high and low flow pulses and their duration in the two stages of dry season and flood season, totaling 8 parameters; the fifth component adds the average flow increase rate, average flow decrease rate, and flow reversal times in the four stages of dry season, pre-flood season, main flood season, and post-flood season, totaling 12 parameters. The parameters of the conventional and improved hydrological variation index systems are shown in Table 2.

[0121] Table 2 Conventional and improved hydrological variation index systems

[0122]

[0123]

[0124] In order to verify the necessity of improving the hydrological variation index system, the parameters under the original statistical scale (full year) and the parameters under the newly added statistical scale (dry season, pre-flood season, main flood season, post-flood season) were compared before and after the database was built. The results are shown in Figure 2. Figure 3 Among them, the arrow represents the changing trend of the parameter, and the double arrow represents a significant changing trend.

[0125] from Figure 3 It can be seen that the change trends of the parameters of the second group, the maximum and minimum one-day flow, the maximum and minimum three-day flow, and the minimum seven-day flow, under the newly added statistical scale (dry season, pre-flood season, main flood season, post-flood season) are basically consistent with the original statistical scale (full year), and all have a downward trend. In addition to the maximum three-day flow in the post-flood season, the other parameters have passed the 95% significance test and the downward trend is significant. For the parameter of the maximum seven-day flow, it shows a significant downward trend on the annual statistical scale, but this rule is only retained in the pre-flood season and the main flood season. Although there is a downward trend in the post-flood season, it is not significant, and it has an opposite upward trend in the dry season.

[0126] The changing trends of the high / low flow pulse number parameters in the fourth group at different statistical scales are also significantly different. With the flattening of the flow process, the high and low flow pulse numbers at the annual scale have a significant downward trend, but the high flow pulse number mainly decreases in the flood season, while the high flow pulse number in the dry season has an upward trend. Similarly, the low flow pulse number mainly decreases in the dry season, while it has an upward trend in the flood season.

[0127] The change trends of the parameters of average flow reduction rate / average flow increase rate / flow reversal number in the fifth group at different scales are basically consistent. Except for the average flow reduction rate in the dry season / post-flood season, the average flow increase rate in the first flood season / main flood season / post-flood season, and the flow reversal number in the post-flood season, the other parameters all passed the 95% significance test. However, there are also large differences in the amount of change of the parameters at different statistical scales.

[0128] In summary, a single statistical scale ignores the seasonal differences of parameters in different statistical periods, resulting in deviations in the trend judgment of parameters, which in turn affects the results of hydrological situation analysis. Therefore, it is necessary to add different statistical scales and improve the conventional hydrological variation indicator system to reflect the seasonal characteristics of each parameter.

[0129] Based on the conventional and improved hydrological variation index systems, the RVA method and HMA method were used to quantitatively calculate the changes in the overall hydrological situation of Liujiaping River before and after the reservoir was built. The results are shown in Figure 4 .

[0130] Under the RVA method, among the 33 hydrological parameters of the conventional hydrological variation index system, only IHA3 (average flow in March), IHA8 (average flow in August), IHA 19 (maximum 30-day flow), IHA 26 (Time when minimum flow occurs), IHA 33 (Flow turnover times) 5 parameters are low-level changes, 21 parameters are high-level changes, and 13 parameters have a change rate of 100%. The overall hydrological change rate has reached 80.26%, which is a high change. Among the 44 new hydrological parameters added to the improved IHA index system, only IIHA 25 (maximum 3-day flow rate during dry season), IIHA 61 (low flow pulse number during flood season), IIHA 68 (Flow reversal times during dry season), IIHA 70 (average increase rate of flow in the pre-flood season), IIHA 73 (average increase rate of flow during the main flood season), IIHA 74 (number of flow reversals during the main flood season), IIHA 77(Number of flow reversals in the post-flood season) 7 parameters are of low change, 32 parameters are of high change, of which 19 parameters have a change rate of 100%. The overall hydrological change rate has reached 81.10%, which is also a high change.

[0131] In addition, the parameter IHA of the second component of the conventional hydrological variation index system 13 ~IHA 18 The change of (maximum and minimum 1-day, 3-day, 7-day flow) is above 90%, and the corresponding hydrological parameter IIHA is added 24 ~IIHA 46 In addition to IIHA 45 (maximum 7-day flow in the post-flood season) is moderately changed, and the other parameters are also highly changed. The original parameters of the fourth component IHA 27 and IHA 28 The changes in (number and duration of high flow pulses) are 90% and 100% respectively, and the corresponding hydrological parameter IIHA is added 53 ~IIHA 56 The parameters IHA 29 and IHA 30 (Number and duration of low flow pulses) are highly variable, but the corresponding hydrological parameter IIHA is newly added 59 ~IIHA 62 Medium, with dry period low flow duration IIHA 60 And the number of low flow pulses during flood season IIHA 61 The two parameters are moderately changed and slightly changed respectively. The original parameter IHA of the fifth component 31 ~IHA 33 The changes in (average flow reduction rate, average flow increase rate and flow reversal times) are small, all below moderate changes. At the same time, the corresponding hydrological parameters IIHA are added. 66 ~IIHA 77 In addition to the average reduction rate of flow in the previous flood season, IIHA 69 and the average reduction rate of post-flood season flow IIHA 75 The changes of two parameters are relatively high, while the changes of the remaining parameters are relatively small.

[0132] In general, the newly added hydrological parameters and the original corresponding hydrological parameters have basically the same degree of change due to their similar construction methods, but the change degrees of some parameters are still inconsistent. Therefore, in order to comprehensively characterize the changing characteristics of the hydrological regime of the Liujiaping River before and after the reservoir construction, it is necessary to improve the conventional IHA index system.

[0133] Under the HMA method, the parameter change value is significantly smaller than that of the RVA method. Among the 33 hydrological parameters in the conventional hydrological variation index system, 7 parameters are low-change, 13 parameters are highly changed, and the number of highly changed parameters is significantly reduced. There are only 6 parameters with a change of more than 90%, and only IHA 23 The change degree of (baseflow index) reached 100%, and the overall hydrological change degree was 64.55%. Among the 44 new hydrological parameters added to the improved IHA index system, 15 parameters were low-change, 16 parameters were highly changed, 10 parameters had a change degree of more than 90%, and no parameter had a change degree of 100%. The overall hydrological change degree was 62.82%.

[0134] Since the RVA method only calculates the degree of change based on whether the parameter falls into the target interval, when the parameter is slightly disturbed and jumps out of the interval, the value of the degree of change will increase significantly. Therefore, the value of the degree of change calculated by the RVA method is too large, and there will be more high-degree of change parameters. HMA comprehensively considers the changes in parameters in different intervals, and it is difficult to have a degree of change of 100%. The calculated value of the degree of change is smaller than that of the RVA method, and the high-degree of change parameters are also reduced to a certain extent. However, no matter which method is used, it shows that due to the construction and dispatching of the reservoir, the hydrological situation of the river downstream of the power station has changed dramatically.

Claims

1. A method for analyzing hydrological changes in small and medium-sized rivers based on an improved hydrological variation index system, characterized in that: The following processes are included: S1: Based on the seasonal characteristics of runoff in small and medium-sized rivers, the hydrological period is divided into four periods: dry season, pre-flood season, main flood season, and post-flood season. S2: Count the hydrological parameters in each period, supplement the parameters of the conventional hydrological variation index system, and propose an improved hydrological variation index system; S3: Based on the improved hydrological variation index system, the variation range evaluation method and histogram matching method are used to calculate the degree of change in the hydrological situation of small and medium-sized rivers.

2. According to claim 1, a method for analyzing hydrological changes in small and medium-sized rivers based on an improved hydrological variation index system is characterized in that: The hydrological stage division adopted by S1 includes the construction of stage indicator set, membership calculation and flood season division; The construction of the staging index set includes staging scale determination and staging index selection; the staging scale is determined with ten days as the scale; The stage index selection adds the minimum and maximum 1-day, 3-day, 5-day, and 7-day runoff indicators based on the average runoff value, constructs the hydrological stage index set X, and uses the CRITIC method to calculate the weight W of each stage index j , transform the stage index set X into the stage comprehensive index set Y; Where p is the number of years of runoff data; X i is the set of indicators for the i-th year; m is the number of indicators selected for each decade (including the average runoff within the decade, the minimum and maximum runoff in 1 day, 3 days, 5 days, and 7 days); is the jth indicator in the kth decade of the i-th year; Y i is the comprehensive indicator set for the i-th year; is the comprehensive index of the kth decade in the i-th year; The membership calculation includes threshold construction and membership definition; the threshold construction is to take weighted average of the mean values ​​of each index as the dividing value between the flood season and the dry season; in, is the mean value of the jth indicator; The membership degree is defined in the following steps: For a given runoff sequence in year i, select the comprehensive indicator The starting period is greater than the threshold L and end of the decade As the beginning and end of the flood season, the series data of p years can be statistically obtained into p flood season intervals Where i = 1, 2, ..., p; Count the number n of each decade falling in p flood season intervals as its membership degree to the flood season; l k =n / p (4) The flood season division includes the division between the flood season and the dry season, and the division within the flood season; The division of flood season and dry season, select λ k = 0.5 as the cutoff level, when λ k >0.5, the ten-day period k is classified as the flood season; when λ k When <0.5, the ten-day period is classified as dry season; The flood season is divided into the pre-flood season, the main flood season and the post-flood season by using the mean change point analysis method. The steps are as follows: The ten-day period with the largest membership degree to the flood season m As the demarcation point, the membership sequence {λ k } is divided into the pre-sequence {λ f ,λ f+1 ,...,λ m } and the subsequent sequence {λ m ,λ m+1 ,...,λ f+d }; where f and d represent the start and length of the flood season; Assume that the previous sequence {λ f ,λ f+1 ,...,λ m The change point of} is the αth decadeλ α , calculate the difference β between the statistics D and D'; β=DD' (6) in, are the sequences {λ f ,λ f+1 ,...,λ α }、{λ α+1 ,λ α+1 ,...,λ m } and {λ f ,λ f+1 ,...,λ m }mean; The change point α traverses from f to m, and the decade s corresponding to the maximum β is selected as the dividing point between the pre-flood season and the main flood season; The change point α traverses from m to f+d, and the decade e corresponding to the maximum β is selected as the dividing point between the main flood season and the post-flood season; the pre-flood season, main flood season and post-flood season correspond to [f,s], [s+1,e] and [e+1,f+d] respectively.

3. The method for analyzing the hydrological situation changes of small and medium-sized rivers based on the improved hydrological variation index system according to claim 1 is characterized in that: The improved hydrological variation index system includes parameters of five components. The differences between the parameters of each component and the conventional hydrological variation index system are as follows: (1) The second component adds the maximum and minimum flow rates for 1 day, 3 days, and 7 days in the four stages of dry season, pre-flood season, main flood season, and post-flood season, adding a total of 24 parameters; (2) The fourth group includes the number and duration of high and low flow pulses in the dry season and flood season, adding a total of 8 parameters; (3) The fifth component adds the average flow increase rate, average flow decrease rate and flow reversal times in the four stages of dry season, pre-flood season, main flood season and post-flood season, adding a total of 12 parameters; (4) The first and third components remain unchanged.

Images