Menu Expand
Stochastic Water Demand Modelling

Stochastic Water Demand Modelling

Mirjam Blokker

(2011)

Additional Information

Book Details

Abstract

Water quality processes in the drinking water distribution network are strongly influenced by the flow velocity and residence time of the water in the network. In order to understand how the water quality changes in the drinking water distribution network, a good understanding of hydraulics is required. Specifically in the periphery of the network, where customers are connected, the hydraulics can change rapidly. During the night time the water is almost stagnant and the residence time increases. In the morning, when everybody gets up and flushes the toilet and takes a shower, high flow velocities can occur. During the remainder of the day flow velocities are low. The stochastic endues model SIMDEUM was developed to simulate water use on a small time scale (1 s) and small spatial scale (per fixture). SIMDEUM enables a good model of flow velocities, residence times and the connected water quality processes in the water distribution network. 
Stochastic Water Demand Modelling: Hydraulics in Water Distribution Networks describes the requirements of hydraulics in water quality modelling and provides insight into the development of detailed residential and non-residential water demand models. The book illustrates the use of detailed demand models in water quality models with respect to the variation in residence times and the relation with particle accumulation and resuspension. The models are compared to measurements in several real drinking water distribution networks. 

Table of Contents

Section Title Page Action Price
Cover page 1
Haif title page 2
Title page 4
Copyright page 5
Contents 6
Chapter 1 12
1.1 WATER DEMAND MODELLING 12
1.2 RESEARCH OBJECTIVES 13
1.3 OUTLINE 13
Chapter 2 16
ABSTRACT 16
2.1 INTRODUCTION 16
2.2 WATER QUALITY MODELLING - DISSOLVED MATTER 18
2.3 WATER QUALITY MODELLING - PARTICULATE MATTER 20
2.4 DEMANDS IN HYDRAULIC NETWORK MODELS 23
2.5 DISCUSSION 26
2.6 CONCLUSIONS 28
Chapter 3 30
ABSTRACT 30
3.1 INTRODUCTION 30
3.2 METHODS AND MATERIALS - STATISTICAL ANALYSIS 31
3.3 METHODS AND MATERIALS - MODEL DEVELOPMENT 32
3.3.1 Basic model 32
3.3.2 Justification of input sources 33
3.3.3 The end uses (K) 34
3.3.4 Users (j) 34
3.3.5 The frequency of use (F) 36
3.3.6 The pulse intensity (I) 36
3.3.7 The pulse duration (D) 37
3.3.8 The diurnal pattern, time of water use (t) 39
3.4 METHODS AND MATERIALS - THE SIMULATION 41
3.5 METHODS AND MATERIALS - MODEL VALIDATION PARAMETERS 44
3.6 METHODS AND MATERIALS - MODEL VALIDATION 46
3.7 RESULTS 46
3.8 DISCUSSION 48
3.9 CONCLUSIONS 50
Chapter 4 52
ABSTRACT 52
4.1 INTRODUCTION 52
4.2 METHODS AND MATERIALS - MODEL DEVELOPMENT 53
4.2.1 Basic model 53
4.2.2 The functional rooms (h) 54
4.2.3 The end uses (K) 54
4.2.4 Users (j) 56
4.2.5 The frequency of use (f) 57
4.2.6 The pulse intensity (I) and pulse duration (D) 57
4.2.7 The diurnal pattern, time of water use (t) 57
4.3 METHODS AND MATERIALS - THE SIMULATION 58
4.4 METHODS AND MATERIALS - MODEL VALIDATION 61
4.5 METHODS AND MATERIALS - SENSITIVITY ANALYSIS 62
4.6 RESULTS - MODEL VALIDATION 62
4.6.1 Office building 62
4.6.2 Hotel 62
4.6.3 Nursing home 65
4.7 RESULTS - SENSITIVITY ANALYSIS 65
4.8 DISCUSSION 65
4.9 CONCLUSIONS 68
Chapter 5 70
ABSTRACT 70
5.1 INTRODUCTION 70
5.2 MATERIALS AND METHODS 71
5.2.1 The Poisson Rectangular Pulse (PRP) model 71
5.2.2 The End-Use Model SIMDEUM 73
5.2.3 The flow measurements of Milford, Ohio 75
5.2.4 Comparing the two models 77
5.2.5 Parameters to compare measurements and simulation results 77
5.3 RESULTS: COMPARING THE TWO MODELS 78
5.3.1 Comparing the two models on underlying principles 78
5.3.2 Comparing the black box models 81
5.3.3 Comparing the input data 82
5.3.4 Comparing the output data: results for single homes 83
5.3.5 Comparing the output data: results for sum of 20 homes 86
5.4 DISCUSSION 87
5.4.1 Practicality 87
5.4.2 Performance 89
5.4.3 Application 90
5.5 CONCLUSION 90
Chapter 6 92
ABSTRACT 92
6.1 INTRODUCTION 92
6.2 METHODS AND MATERIALS - BENTHUIZEN AREA 94
6.2.1 Generic methodology 94
6.2.2 The network 94
6.2.3 Measurement setup for the tracer study 95
6.2.4 Hydraulic model and demand allocation 97
6.2.5 Water demand pattern generation 98
6.2.6 Water quality model 99
6.2.7 Sensitivity analysis and model validation 99
6.3 RESULTS AND DISCUSSION - BENTHUIZEN AREA 100
6.3.1 Demand multiplier pattern 100
6.3.2 Residence time - sensitivity analysis 102
6.3.3 Residence time - model validation 107
6.4 INTERMEDIATE CONCLUSIONS FROM BENTHUIZEN STUDY 107
6.5 METHODS AND MATERIALS - ZANDVOORT AREA 107
6.5.1 The network 108
6.5.2 Measurement setup for the tracer study 109
6.5.3 Hydraulic model and demand allocation 111
6.5.4 Water demand pattern generation 113
6.5.5 Model validation 114
6.6 RESULTS AND DISCUSSION - ZANDVOORT AREA 114
6.6.1 Diurnal flow pattern 114
6.6.2 Residence time 115
6.6.3 Pulse shape 116
6.7 INTERMEDIATE CONCLUSIONS FROM ZANDVOORT STUDY 118
6.8 GENERAL DISCUSSION 118
6.9 CONCLUSION 120
Chapter 7 122
ABSTRACT 122
7.1 INTRODUCTION 122
7.2 METHODS AND MATERIALS 124
7.2.1 Introduction 124
7.2.2 The network 124
7.2.3 Flushing the network 124
7.2.4 Water demand pattern generation 126
7.2.5 The hydraulic network model and demand allocation 131
7.2.6 Determining the relationship between hydraulics and settled sediment 131
7.3 RESULTS AND DISCUSSION 132
7.3.1 Interpretation of results 132
7.3.2 Theory of self-cleaning 135
7.3.3 Practical implications 136
7.3.4 Design implications for self-cleaning networks 136
7.4 CONCLUSIONS 137
Chapter 8 140
8.1 INTRODUCTION 140
8.2 CONSTRUCTING THE END-USE MODEL SIMDEUM 140
8.2.1 The approach of modelling the end user 140
8.2.2 Advantage of end-use modelling 141
8.2.3 An undemanding model 142
8.2.4 Pressure driven demand 142
8.3 CASE STUDIES - MEASUREMENT ISSUES 143
8.4 CASE STUDIES - NETWORK SOLVER CONSIDERATIONS 144
8.5 EVALUATION OF ADDED VALUE OF SIMDEUM 145
8.5.1 Effects of detailed demand model in DWDS network model 145
8.6 APPLICATIONS OF SIMDEUM IN RESEARCH ON WATER QUALITY IN THE DWDS 151
8.6.1 Fouling prediction tool 151
8.6.2 Design principles for self-cleaning networks 151
8.6.3 Maximum travel time 151
8.6.4 Sensor placement 151
8.7 PRACTICAL APPLICATIONS OF SIMDEUM 152
8.7.1 Water demand management 152
8.7.2 Scenario studies 152
8.7.3 Risk analysis of contamination through ingestion 152
8.7.4 Dimensioning plumbing and water heaters 153
8.7.5 Water-related energy use 153
8.7.6 Practical application of SIMDEUM in network modelling 153
Chapter 9 156
9.1 INTRODUCTION 156
9.2 DEVELOPING AND VALIDATING SIMDEUM: A SIMULATION MODEL FOR DRINKING WATER DEMAND 156
9.3 APPLYING SIMDEUM IN CASE STUDIES: THE STUDY OF RESIDENCE TIMES AND PARTICLES IN THE DISTRIBUTION NETWORK 157
9.4 EVALUATING ADDED VALUE OF SIMDEUM: COMPARING SIMDEUM TO EXISTING MODELS 157
9.5 EVALUATING ADDED VALUE OF SIMDEUM: NEW WAY OF DEMAND MODELLING IS REQUIRED FOR MODELLING WATER QUALITY IN THE DISTRIBUTION N 158
References 160
Appendix 166
A.1 AUTO- AND CROSS-CORRELATION 166
A.2 GOODNESS-OF-FIT 167
A.3 PROBABILITY DISTRIBUTION FUNCTIONS 170
A.4 STATISTICAL TEST FOR NORMAL DISTRIBUTION 170
A.4.1 Introduction 170
A.4.2 Pearson’s χ2-test 171
A.4.3 Lilliefors test 172
REFERENCES 174
LIST OF SYMBOLS 174