diff --git a/python/paddle/fluid/tests/unittests/test_batch_norm_op.py b/python/paddle/fluid/tests/unittests/test_batch_norm_op.py
index c2a6c468e5c8f184e6e74a98151d6c008ba997d4..02171db3fca75a7fbd655771818f2e21fea52f5d 100644
--- a/python/paddle/fluid/tests/unittests/test_batch_norm_op.py
+++ b/python/paddle/fluid/tests/unittests/test_batch_norm_op.py
@@ -768,6 +768,10 @@ class TestBatchNormOpError(unittest.TestCase):
             )
             self.assertRaises(TypeError, paddle.static.nn.batch_norm, x2)
 
+            # the first dimension of input for batch_norm must between [2d, 5d].
+            x3 = paddle.static.data("", shape=[0], dtype="float32")
+            self.assertRaises(ValueError, paddle.static.nn.batch_norm, x3)
+
 
 class TestDygraphBatchNormAPIError(unittest.TestCase):
     def test_errors(self):
diff --git a/python/paddle/fluid/tests/unittests/test_fold_op.py b/python/paddle/fluid/tests/unittests/test_fold_op.py
index 1f3193fa1fd49470678a69a7143eba43dc49b878..a86161cc4502308c61dbde77ffff9c608b40e6ff 100644
--- a/python/paddle/fluid/tests/unittests/test_fold_op.py
+++ b/python/paddle/fluid/tests/unittests/test_fold_op.py
@@ -179,7 +179,7 @@ class TestFoldOpError(unittest.TestCase):
         with program_guard(Program(), Program()):
 
             def test_input_shape():
-                # input_shpae must be 3-D
+                # input_shape must be 3-D
                 x = paddle.randn(shape=[2, 3, 6, 7], dtype="float32")
                 out = fold(x, output_sizes=[2, 3], kernel_sizes=[2, 2])
 
diff --git a/python/paddle/static/nn/common.py b/python/paddle/static/nn/common.py
index 3b40153cbb797d76ace0e3a950e533f2b36d6648..c43385a8e91409f90535ba428d1b83b9715ec503 100644
--- a/python/paddle/static/nn/common.py
+++ b/python/paddle/static/nn/common.py
@@ -2731,6 +2731,12 @@ def batch_norm(
         dtype = core.VarDesc.VarType.FP32
 
     input_shape = input.shape
+    if len(input.shape) < 2 or len(input.shape) > 5:
+        raise ValueError(
+            'expected 2D or 3D or 4D or 5D input (got {}D input, input shape is: {})'.format(
+                len(input.shape), input_shape
+            )
+        )
     if data_layout == 'NCHW':
         channel_num = input_shape[1]
     else: